A relationship between separation anxiety disorder (SAD)in children and panic disorder (PD) in adults has been sug-gested. Klein (1964) first proposed the “separation anxietyhypothesis” of panic disorder on the basis of clinical retro-spective reports of increased rates of childhood separationanxiety symptoms in adults with agoraphobia and panicattacks, as compared with other psychiatric patients. Althoughnot all studies have been consistent (Biederman et al., 2001;Lipsitz et al., 1994), several studies provide converging

evidence of a relationship between SAD and PD. Retro-spective studies describe increased rates of SAD in adultswith PD (Ayuso et al., 1989; Balon et al., 1989; Zitrinand Ross, 1988). A prospective longitudinal study of chil-dren with SAD reports an elevated rate of PD among thesechildren in adulthood (Klein, 1995). In addition, familystudies report increased rates of SAD in offspring of par-ents with PD and agoraphobia (Berg, 1976; Capps et al.,1996; Weissman et al., 1984). A relationship between SADand PD is also suggested by psychophysiological studies.In a study of carbon dioxide sensitivity, Pine and colleagues(2000) found patterns of respiratory abnormalities specif-ically in children with SAD, but not social anxiety disor-der, that were similar to those found in adults with PD(Papp et al., 1997).

Allergic disorders provide an additional framework inwhich to examine the association between SAD and PD.Specifically, a relationship has been suggested betweenPD in adults, anxiety disorders in children, and a famil-ial cluster of IgE-mediated allergies often referred to as

Relationship Between Separation Anxiety Disorder,Parental Panic Disorder, and Atopic Disorders in

Children: A Controlled High-Risk Study

MARCIA J. SLATTERY, M.D., M.H.S., DONALD F. KLEIN, M.D., SALVATORE MANNUZZA, PH.D., JOHN L. MOULTON III, PH.D., DANIEL S. PINE, M.D., AND RACHEL G. KLEIN, PH.D.

ABSTRACT

Objective: To test the hypotheses that rates of atopic disorders are elevated in offspring of parents with panic disorder

(PD) and in children with separation anxiety disorder (SAD). Method: Rates of atopic disorders were assessed in 343

offspring (aged 6–17 years) of parents with PD, nonpanic psychiatric disorders, and no psychiatric disorder. Lifetime his-

tory of atopic disorders was determined by parental responses to a clinician-administered questionnaire assessing med-

ical treatment for asthma and allergies. Logistic regression analyses assessed the association between atopic disorders

and parental PD, and between atopic disorders and probable or definite childhood SAD. Analyses controlled for age,

sex, socioeconomic status, and treatment for other medical illnesses. Results: Increased rates of atopic disorders were

found in offspring of parents with PD (odds ratio [OR] = 2.56, 95% confidence interval [CI] = 1.27–5.16, p = .009) and in

children with SAD (OR = 2.71, 95% CI = 1.22–6.03, p = .015). Associations remained significant when both parental PD

and SAD were included in the model, suggesting that each contributed independently to increased rates of atopy.The inter-

action of parental PD and child SAD was not significant. Conclusions: Atopic disorders in children are associated with

parental PD and with childhood SAD. Results do not appear to support that having both childhood SAD and a parent

with PD confers increased risk for atopic disorders above and beyond either condition alone. J. Am. Acad. Child Adolesc.

Psychiatry, 2002, 41(8):947–954. Key Words: separation anxiety disorder, panic disorder, atopy, asthma, allergy.

J . AM. ACAD. CHILD ADOLESC. PSYCHIATRY, 41 :8 , AUGUST 2002 947

Accepted March 8, 2002.Dr. Slattery is with the Mayo Clinic, Rochester, MN, and the New York

University Child Study Center, New York; Dr. D.F. Klein is with the New YorkState Psychiatric Institute, New York; Drs. Mannuzza, Moulton, Pine, and R.G.Klein are with the New York University Child Study Center.

This research was supported by U.S. Public Health Service grants MH-37592and MH-30906. The authors thank Mary Guardino, Director, Freedom From Fear,Staten Island, NY, for access to patients, and Maren K. Olsen, Ph.D., Bioinformationand Biostatistics Department, Duke University, for review of analyses.

Reprint requests to Dr. Slattery, Mayo Clinic, 200 First Street S.W., Rochester,MN 55905.

0890-8567/02/4108–0947�2002 by the American Academy of Childand Adolescent Psychiatry.

atopy (Schmidt-Traub and Bamler, 1997). The term moreaccurately describes an inherited predisposition to thedevelopment of these disorders, rather than the disordersthemselves (Panhuysen et al., 1995). Atopic disordersinclude asthma, urticaria (hives), allergic rhinitis (hayfever), and atopic dermatitis (eczema) (Blumenthal, 1997).

Several studies provide evidence of a relationshipbetween asthma, allergies, and PD in adults. Clinicalstudies of adults with asthma report increased rates ofPD as compared with the prevalence of PD in commu-nity studies (Perna et al.,1997; Shavitt et al., 1992; Yellowleeset al., 1988). Conversely, clinical studies of adults withPD describe increased rates of respiratory disease, includ-ing asthma, as compared with obsessive-compulsive dis-order, eating disorders, and V-code diagnoses (Perna et al.,1994; Spinhoven et al., 1994; Zandbergen et al., 1991).Schmidt-Traub and Bamler (1997) examined the rate ofPD in subjects with IgE-mediated allergies (n = 100) andthe rate of allergies in subjects with PD (n = 79); eachgroup was compared with a healthy control group (n =66). The rate of PD (10%) and panic attacks (35%) insubjects with allergies was elevated compared with ratesof PD (2%) and panic attacks (17%) in controls. Subjectswith PD had an increased rate of allergies (70%) com-pared with controls (29%).

There is a paucity of clinical and epidemiological stud-ies examining the relationship between allergic disordersand specific childhood anxiety disorders. Several studieshave suggested a relationship between asthma and anxi-ety symptoms in children but did not assess specific anx-iety diagnoses (Butz and Alexander, 1993; Graham et al.,1967; McNichol et al., 1973; Norrish et al., 1977; Silvergladeet al., 1994). Of the studies that used structured diag-nostic interviews to examine specific anxiety disorders inasthmatic children, results differ as to whether SAD orgeneral anxiety disorder (GAD) occurs most often. Vilaand colleagues (2000) reported rates of anxiety disordersin 82 children with asthma (mean age = 11); 35% hadany anxiety disorder, 29% had GAD, 16% had SAD,10% had social anxiety disorder, and 1% had PD. In asimilar study of 62 asthmatic children (mean age = 15),Wamboldt and colleagues (1996) described rates of anx-iety disorders including any anxiety disorder (28%), over-anxious disorder (24%), SAD (15%), agoraphobia (3%),simple phobia (6%), and PD (0%). Bussing and col-leagues (1996) compared rates of anxiety in 37 asthmaticchildren (mean age = 11) compared with 31 healthy con-trols. All anxiety disorders were elevated in children with

asthma as compared with controls, including any anxi-ety disorder (43% versus 19%), SAD (32% versus 13%),overanxious disorder (8% versus 0%), PD (5% versus0%), and phobia (16% versus 6%).

The prevalence of specific anxiety disorders in chil-dren with atopic disorders other than asthma is unknown.Similarly, there are no published studies investigating therate of atopic disorders in clinical samples of childrenwith anxiety disorders. Indirect support for a relation-ship between atopy and anxiety disorders among chil-dren may be suggested by other data. Kagan (1994)reported increased rates of allergies in children with behav-ioral inhibition. This may suggest a relationship betweenatopy and anxiety disorders in children, since behavioralinhibition is a temperamental trait associated with increasedrates of anxiety in children (Hirshfeld et al., 1992).

Existing studies do not provide insight into potentialmechanisms for the association between anxiety and atopicdisorders. Wamboldt and colleagues (1998) suggested thatatopic disorders may represent markers for a familial sub-type of anxiety characterized by genetic factors shared withatopy. If this is correct, an anxiety disorder in the parentwould predict the presence of asthma or allergies in thechild, even in the absence of anxiety in the child. Preliminaryevidence for this model is suggested in one study byMerikangas and colleagues (1999), who reported increasedrates of allergies and eczema in children of parents withanxiety disorders compared with children of parents withsubstance abuse or no psychiatric disorder.

The present study further examines these associationsby investigating the relationship between atopy, parentalPD, and SAD in children. We hypothesize that (1) off-spring of parents with PD will have increased rates ofatopic disorders as compared with offspring of parentswithout panic disorder, and (2) that children with SADwill have increased rates of atopic disorders as comparedwith children without SAD.

METHOD

Subjects

Data were drawn from a larger study examining the risk for psy-chopathology conferred to children by parents with anxiety and mooddisorders compared with non-ill parents. Nonpsychotic psychiatri-cally ill parents with anxiety and mood disorders with children aged6 to 17 years were identified by random chart review from three out-patient treatment centers in the New York City area. Non-ill com-parison parents were identified by chart review and the acquaintanceshipprocedure. Charts of a major New York medical center dental clinicwere reviewed to identify children aged 6 to 17 years. Identified par-

SLATTERY ET AL.

948 J . AM. ACAD. CHILD ADOLESC. PSYCHIATRY, 41 :8 , AUGUST 2002

ents were screened by a social worker who administered a brief psy-chiatric questionnaire; those with negative screens were scheduled fora psychiatric interview (described below). In the acquaintanceshipprocedure, parents with anxiety and mood disorders (from the psy-chiatrically ill group) were asked to identify six acquaintances whowere the same age, gender, and ethnicity and who had children aged6 to 17. One acquaintance was randomly chosen and contacted. Aswith the chart review procedure, parents were screened and subse-quently interviewed. Approximately half of the control group wasrecruited through each of the recruitment methods.

There were three parental diagnostic groups. In the PD group (n =95 families, 150 children), one or both parents had lifetime PD andneither parent had lifetime mood or psychotic disorder. In the non-panic anxiety and/or mood disorder group (n = 69 families, 113 chil-dren), one or both parents had lifetime nonpanic anxiety disorderincluding social phobia (n = 12, 17%), obsessive-compulsive disorder(n = 4, 6%), and/or a mood disorder including major depressive dis-order (n = 62, 90%), dysthymia (n = 12, 17%) or bipolar I disorder (n = 8, 12%), and neither parent had lifetime psychotic disorder. Inthe control group (n = 44 families, 80 children), neither parent hadlifetime anxiety, mood, or psychotic disorder.

Subjects were 343 offspring, aged 6 to 17 years (mean = 11.6, SD =3.5). Offspring were grouped on the basis of parental and child psy-chiatric diagnoses.

Written informed consent (and child assent) was obtained from allsubjects, regardless of group or method of recruitment.

Measures

Overview. Interviewers were advanced-level clinical psychology doc-toral students, a clinical psychologist, and a psychiatric social workertrained in differential diagnostic assessment. Interview training includedrequired readings, discussion of differential diagnoses of symptoms anddisorders, extensive review of diagnostic instruments and existing nar-rative summaries, and completion of 40 DSM-IV case vignettes. Ongoingsupervision (to minimize “rater drift”) included regularly scheduledmeetings to receive feedback on narrative summaries and to review diag-nostically ambiguous cases. Interviewers were blind to group member-ship and study hypotheses. Parents and children were assessed by differentinterviewers, who wrote detailed narratives that were blindly reviewedby senior investigators. Narrative reviews served as a validity check toensure that diagnostic criteria were carefully followed.

Parent Diagnoses. Parents were administered the Structured ClinicalInterview for DSM-IV (First et al., 1995; Spitzer et al., 1992), whichassesses lifetime presence of all major psychiatric illnesses. When thesecond parent was not available for interview (49% of cases), theFamily Informant Schedule and Criteria-Updated for DSM-IV (Schleyerand Mannuzza, 1995) was administered to the available parent.

Child Diagnoses. An informant interview with the child’s motherwas conducted; children aged 9 years and older were also intervieweddirectly. Children and mothers were administered the Parent asRespondent Informant Schedule (PARIS; Klein and Mannuzza, 1992;Kentgen et al., 1997). The PARIS assesses lifetime presence of all majorchild and adolescent DSM-IV disorders. Child lifetime diagnoses wereconsidered present if DSM-IV criteria were met in the direct interviewwith the child and/or in the parent interview about the child. A defi-nite disorder was diagnosed when DSM-IV criteria were met; a prob-able diagnosis was assigned when functional impairment was presentdespite fewer symptoms (Gittelman et al., 1985). A disorder was con-sidered present on the basis of either a probable or definite diagnosis.This decision was considered consistent with clinical practice, sinceboth levels of certainty required significant functional disruption.

Atopy. Atopy was assessed with a medical questionnaire adminis-tered to the parents by clinicians. Questionnaire items included treat-ment for asthma, allergy, diabetes, seizures, or any other illness. Thepresence of an atopic disorder was defined as an affirmative responseto at least one of two (yes/no) questions regarding whether the childhad ever received medical treatment for asthma or allergies.

Data Analysis

Atopy was the dependent variable. Because children included sib-lings, variables could not be considered independent. Therefore, asso-ciations between atopy and parental PD and between atopy and SADwere assessed with generalized estimating equations (GEE) using theSAS statistical program (SAS, Version 8.1, SAS Institute Inc., Cary,NC). GEE, an extension of the general linear model, permits theanalysis of clustered binary data that may be correlated. In this case,clusters were the 208 families in the study; the binary outcome wasatopy (yes/no). Age and sex were included as covariates on the basisof increased rates of asthma in prepubertal children, particularly boys(Schachter et al., 1984; Sly, 1996). Because treatment seeking may beassociated with parental and child psychological distress (Garraldaand Bailey, 1986; Horwitz et al., 1985), as well as parental character-istics of higher education and income (Horwitz et al., 1985), we con-trolled for SES and for medical treatment reported by parents for allother illnesses. We calculated odds ratios (OR) and 95% confidenceintervals (CI) to assess the strength of the associations and examinedrobust standard errors to reduce the likelihood of type I errors. Alphawas set at p < .05 for a priori hypotheses and p < .01 for secondaryanalyses. All contrasts were two-tailed. The output of analyses usingGEE and logistic regression (SPSS, Version 10.1.0; SPSS Inc., Chicago,IL) were compared. Because the significance of the contrasts wereidentical between the analyses (i.e., a significant cluster effect was notpresent), we report results from the logistic regression.

RESULTS

Demographic Characteristics

The 343 children ranged in age from 6 years 0 monthsto 17 years 11 months (mean = 11.6 years, SD = 3.5 years)(Table 1). There were slightly more girls (54%) than boys.The majority of children were white (76%), with theminority comprising African American (11%), Hispanic(8%), and other (5%) ethnic groups. Household socioe-conomic status ranged from 1 to 5 (mean = 2.9, SD =1.0), with 1 representing upper class and 5 representinglower class (Hollingshead and Redlich, 1958). There were40 children with treated atopy (12%); 26 (9%) had asthma,10 (4%) had allergies, and 4 (1%) had both asthma andallergies. Paired comparisons of offspring of parentswith/without PD, children with/without SAD, and chil-dren with/without atopy (Table 1) indicated a significantcontrast (p < .05) for age only in offspring of parents withPD versus no PD; all other contrasts were nonsignificant.

The majority of parents were married (67%), with theminority comprising separated (9%), divorced (15%),widowed (1%), and single (8%) parents. Mothers’ meanage was 41.2 years; the mean age of fathers was 44.0 years.

SEPARATION ANXIETY, PARENTAL PANIC, ATOPY

J . AM. ACAD. CHILD ADOLESC. PSYCHIATRY, 41 :8 , AUGUST 2002 949

Prevalence of Atopy in Offspring of Parents Withand Without PD

There was a significantly higher rate of treated atopyamong the offspring of parents with PD as comparedwith offspring of parents without PD (OR = 2.56, 95%CI = 1.27–5.16, p = .009) (Table 2). Secondary analysesexamined the diagnostic specificity of this association.Children were grouped on the presence or absence of par-ent diagnoses for the purpose of conducting two con-trasts. There were too few offspring of parents withnonpanic anxiety disorders (n = 27) to permit compari-son with offspring of parents with PD. Offspring of par-ents with PD were compared with offspring of parentswith other disorders (i.e., any nonpanic anxiety or mooddisorder) and offspring of control parents (i.e., parentswithout any mood, anxiety or psychotic disorders).Offspring of parents with PD had significantly higherrates of atopic disorders compared with offspring of par-ents with other psychiatric disorders (OR = 3.35, 95%CI = 1.37–8.15, p = .008). Although the odds of atopyamong the offspring of parents with PD was nearly twicethe odds among the offspring of control parents, the dif-ference did not reach significance (OR = 1.92, 95% CI =0.78–4.72, p = .15) (Table 2). Finally, rates of atopic dis-orders in children with one or both parents with PD werecompared. Two of six (33%) children with PD in bothparents had atopic disorders, compared with 23 of 144(16%) children with PD in one parent.

Prevalence of Atopy in Children With and Without SAD

Children with probable or definite SAD had a signif-icantly higher rate of atopic disorders than children with-out SAD (OR = 2.71, 95% CI = 1.22–6.03, p = .015)(Table 3). Secondary analyses examined whether this rela-tionship had diagnostic specificity. Children with SADwere compared with children with non-SAD anxiety dis-orders and children with no history of psychiatric illness.Non-SAD anxiety disorders included GAD (n = 9, 8%),PD (n = 2, 2%), social phobia (n = 47, 40%), specific pho-bia (n = 77, 65%), obsessive-compulsive disorder (n = 3,3%), posttraumatic stress disorder (n = 1, 1%), and anx-iety disorder not otherwise specified (n = 5, 4%). Restrictinganalyses to definite (i.e., full criteria) diagnoses of SADand non-SAD anxiety disorders did not alter any of therelationships reported.

The rate of atopic disorders in children with SAD wassignificantly higher than that observed among childrenwho had never been psychiatrically ill (OR = 5.85, 95%CI = 1.79–19.09, p = .003) (Table 3). Although the oddsof atopic disorders among children with SAD were twiceas high as those observed among children with non-SADanxiety disorders, the difference did not reach signifi-cance (OR = 2.14, 95% CI = 0.86–5.37, p = .10).

Further analyses examined the relationship betweenparental PD and childhood SAD as related to childhoodatopy. The association between atopic disorders andparental PD, and between atopic disorders and child SAD,

SLATTERY ET AL.

950 J . AM. ACAD. CHILD ADOLESC. PSYCHIATRY, 41 :8 , AUGUST 2002

TABLE 1Child Demographic Variables (N = 343 Children)

Children With Offspring of Parents

All ChildrenSeparation Anxiety Disorder a With Panic Disorder * Children With Atopy a

Yes No Yes No Yes No(N = 343) (n = 49) (n = 294) (n = 150) (n = 193) (n = 40) (n = 303)

Gender, n (%)Female 186 (54) 19 (39) 138 (47) 72 (48) 85 (44) 23 (58) 134 (44)Male 157 (46) 30 (61) 156 (53) 78 (52) 108 (56) 17 (43) 169 (56)

Ethnicity, n (%)White 262 (76) 33 (67) 229 (78) 119 (79) 143 (74) 28 (70) 234 (77)AfAm 38 (11) 9 (18) 29 (10) 10 (7) 28 (15) 5 (13) 33 (11)Hispanic 28 (8) 4 (8) 24 (8) 17 (11) 11 (6) 4 (10) 24 (8)Other 15 (4) 3 (6) 12 (4) 4 (3) 11 (6) 3 (8) 12 (4)

Age, mean (SD) 11.6 (3.5) 11.4 (3.3) 11.5 (3.5) 10.9 (3.3) 11.9 (3.5) 12.0 (3.8) 11.4 (3.4)[Range] [6.0–17.9] [6.0–17.6] [6.0–17.9] [6.0–17.6] [6.0–17.9] [6.2–17.9] [6.0–17.9]

SES, mean (SD) 2.9 (1.0) 3.1 (1.1) 2.9 (1.0) 2.9 (1.0) 2.9 (1.1) 2.9 (1.0) 2.9 (1.1)

Note: AfAm = African American; SES = socioeconomic status.a No significant group differences.* Significant difference only for age, PD (yes) < PD (no), t341 = 2.61; p < .05.

SEPARATION ANXIETY, PARENTAL PANIC, ATOPY

J . AM. ACAD. CHILD ADOLESC. PSYCHIATRY, 41 :8 , AUGUST 2002 951

TABLE 2Relationship Between Parental Panic Disorder and Atopic Disorders in Offspring

Contrasts of Ill-Parent Diagnostic Groupsχ2

Parental Diagnoses Adjusted OR (95% CI)

A BParental Panic Disorder: Yes Parental Panic Disorder: No

(n = 150 Offspring) (n = 193 Offspring) A vs. B

Atopy,a n (%) χ21 = 6.84; p = .01

OR = 2.56Yes 25 (16.7) 15 (7.8) (1.27–5.16)No 125 (83.3) 178 (92.2) p = .009

BNonpanicAnxietyDisorder C

A and/or Mood Not IllPanic Disorder Disorderb Controls

(n = 150) (n = 113) (n = 80) A vs. B A vs. C B vs. C

Atopy,a n (%) χ21 = 5.40; p = .02 χ2

1 = 2.73; p = .10 χ21 = 0.18; p = .67

OR = 3.35 OR = 1.92 OR = 0.56Yes 25 (16.7) 8 (7.1) 7 (8.7) (1.37–8.15) (0.78–4.72) (0.17–1.80)No 125 (83.3) 105 (92.9) 73 (91.3) p = .008 p = .15 p = .33

Note: Logistic regression controlled for age, sex, socioeconomic status, and treatment for other medical illnesses. OR = odds ratio; CI = con-fidence interval.

a Atopy = lifetime history of allergy or asthma in the child.b DSM-IV nonpanic anxiety and/or mood disorders in parents include social phobia, obsessive-compulsive disorder, major depressive disorder,

dysthymia, and bipolar I disorder.

TABLE 3Relationship Between Separation Anxiety Disorder (SAD) and Atopic Disorders in Offspring

Contrasts of Ill-Child Diagnostic Groupsχ2

Child Psychiatric Disorder Adjusted OR (95% CI)

A BSAD: Yes SAD: No

(n = 49 Children) (n = 294 Children) A vs. B

Atopy,a n (%) χ21 = 6.46; p = .01

OR = 2.71Yes 11 (22.4) 29 (9.9) (1.22–6.03)No 38 (77.6) 265 (90.1) p = .015

BNon-SAD

A Anxiety CSAD Disorderb Never Ill

(n = 49) (n = 118) (n = 105) A vs. B A vs. C B vs. C

Atopy,a n (%) χ21 = 3.05; p = .08 χ2

1 = 11.23; p = .001 χ21 = 3.60; p = .06

OR = 2.14 OR = 5.85 OR = 2.23Yes 11 (22.4) 14 (11.9) 5 (4.8) (0.86–5.37) (1.79–19.09) (0.74–6.75)No 38 (77.6) 104 (88.1) 100 (95.2) p = .10 p = .003 p = .155

Note: Logistic regression controlled for age, sex, socioeconomic status, and treatment for other medical illnesses. OR = odds ratio; CI = con-fidence interval.

a Atopy = lifetime history of allergy or asthma in the child.b DSM-IV non-SAD anxiety disorders include social phobia, obsessive-compulsive disorder, generalized anxiety disorder, specific phobia,

posttraumatic stress disorder, and anxiety disorder-not otherwise specified. “B” comparison group did not include 71 children with psychiat-ric disorders other than anxiety disorders.

remained significant when both variables were includedin the logistic regression model (OR = 2.32, 95% CI =1.14–4.72, p = .02; OR = 2.31, 95% CI = 1.02–5.24,p < .05, respectively). However, the interaction of parentalPD and child SAD was not significant (OR 0.37, 95%CI = 0.07–1.93, p = .24) in predicting atopy, indicatingthat the effects of parental PD and child SAD were notmultiplicative. The proportion of children with atopicdisorders with/without parents with PD and with/without SAD were compared (Table 4). Offspring of par-ents with PD who also had SAD had the highest rate ofatopic disorders (24%); offspring of parents without PDor SAD had the lowest rate (6%). Analyses of variancefor proportions with arcsine transformation showed a sig-nificant main effect for parental PD and SAD, but theinteraction was not significant.

DISCUSSION

We report on the relationship of atopic disorders inchildren with parental PD and childhood SAD in a well-defined sample, using standardized measures. There wasa specific association between atopic disorders and parentalPD compared with other parental psychiatric disorders.The lack of statistical significance between rates of atopyin offspring of parents with PD compared with offspringof control parents may be related to the smaller numberof offspring in the parent control group (n = 80) com-pared with the number of offspring in the parent non-PD anxiety and/or mood disorder group (n = 113). Ratesof atopic disorders were also significantly higher in chil-dren with SAD relative to children without SAD, andmore specifically, in contrast to never-ill children. Additionalstudies are necessary to assess the relationship betweenatopic disorders in children with SAD and other types ofanxiety disorders.

Finally, parental PD and child SAD each appear to con-tribute independently to increased rates of atopic disorders

among children. The results of the present study do notsuggest a multiplicative effect of parental PD and child SADin predicting atopy among children. That is, the risk ofatopic disorders in children who are offspring of parentswith PD and who have SAD does not appear to be aboveand beyond that of either condition alone. The elevatedrate of atopic disorders in children with PD in both par-ents (33%) compared with that of children with PD in oneparent (16%) and no PD in either parent (8%) may reflecta genetic relationship between parental PD and atopy.However, due to small sample sizes, this finding must beviewed as suggestive only and requires further study.

The present study adds to the scant data on the rela-tionship between atopy and anxiety disorders in children.It is the first to report a relationship between parental PDand childhood atopic disorders in a high-risk sample.Merikangas et al. (1999) reported an increased rate of aller-gies in offspring of parents with anxiety disorders com-pared with offspring of parents with no psychiatric disorderor parents with a history of substance abuse and no anx-iety disorder; whether this relationship is specific to parentalPD is not reported. The present study supports earlierreports of a relationship between asthma and SAD andsuggests that the association includes allergies as well. Toour knowledge, no other clinical studies have examinedatopy in children with specific anxiety disorders.

The relationship between atopy and parental PD, andbetween atopy and SAD, may be due to several factors.Shared genetic and/or environmental variables may explainsome or all of the observed association. Wamboldt andcolleagues (1998) suggest that anxiety and atopic disor-ders may reflect an additive genetic effect. If so, prelim-inary findings from the present study are consistent witha genetic association between atopy and SAD, and atopyand PD. Alternatively, the association of atopic disorderswith PD and SAD may reflect shared environmental deter-minants. For example, some studies suggest an associa-tion between family dysfunction and increased levels ofanxiety in children with asthma (Bender et al., 2000;Bussing and Burket, 1993). Other studies suggest thatchronic illnesses such as asthma may invoke increasedlevels of fear, worry, and overprotectiveness in parents,thus contributing to separation anxiety in the child(Bussing et al., 1996; Mrazek et al., 1998).

In addition, the co-occurrence of atopic disorders andanxiety disorders may reflect similar mechanisms of neu-robiological dysfunction. Respiratory studies of panicdisorder and asthma in adults propose a model of cen-

SLATTERY ET AL.

952 J . AM. ACAD. CHILD ADOLESC. PSYCHIATRY, 41 :8 , AUGUST 2002

TABLE 4Rate of Atopic Disorders Among Children With/Without SAD

and With/Without Parental PD

Parental PD

Child SAD Yes No

Yes 7/29 (24%) 4/20 (20%)No 18/121 (15%) 11/173 (6%)

Note: Numerators sum to 40 = total number of children with atopicdisorders. SAD = separation anxiety disorder; PD = panic disorder.

tral CO2 chemoreceptor hypersensitivity and activationof a suffocation alarm system (Klein, 1993; Papp et al.,1993). Similar respiratory mechanisms may explain therelationship between childhood asthma and SAD, as sug-gested by preliminary evidence of respiratory parallelsbetween PD and SAD (Pine et al., 2000). Immune dys-function may mediate physiological changes in childrenwith atopic disorders and anxiety disorders, as suggestedby abnormalities in cell-mediated and humoral immu-nity in adults with PD (Andreoli et al., 1992; Coplanet al., 1999; Ramesh et al., 1991; Rapaport, 1998). Finally,similarities in autonomic and central nervous systemadrenergic and cholinergic dysfunction are suggested instudies of allergy (Lemanske et al., 1985; Marshall,1993),adult PD (Noyes et al., 1992; Weissman et al., 1990;Yeragani et al., 1991), SAD (Rogeness et al., 1990), andchildhood asthma (Miller and Wood, 1997).

Limitations

Although children were selected from a well-definedgroup of parents, results may not generalize to a commu-nity sample. For example, severity of parental psychopa-thology, method of recruitment, and procedure for diagnosticassessment may be relevant variables that may differ in acommunity sample. Also, low base rates of atopic disor-ders and small numbers of children in each anxiety group,especially GAD, prevented the comparison of rates of atopicdisorders between “pure” disorders. Determination of atopywas based on rates of medically treated asthma or allergies.Medical testing (e.g., skin tests) combined with clinicalhistory would provide more objective evidence of atopicstatus. Finally, offspring of parents with PD and childrenwith SAD may have increased somatic concerns, increasedhealth care use, and hence, increased opportunity for med-ical detection and treatment of atopic disorders comparedwith other psychiatric groups. We attempted to addressthese potential differences by controlling for other treatedillnesses in the logistic regression.

Clinical Implications

The association between childhood atopic disordersand anxiety disorders in children and their parents high-lights the importance of a medical and psychiatric his-tory in the clinical care of children. Co-occurring anxietyand atopic disorders may affect the diagnosis and man-agement of one another, as suggested in studies of symp-tom severity and use of medications in children withasthma and anxiety (ten Thoren and Petermann, 2000).

Furthermore, awareness of the association of atopic dis-orders and parental PD could result in earlier identifica-tion and treatment of children with allergies. Finally, theassociation between atopic disorders and anxiety disor-ders in children provides a medical model for furtherinvestigation of neurobiological mechanisms associatedwith childhood anxiety disorders.

REFERENCES

Andreoli AA, Keller SE, Rabaeus M, Zaugg L, Garrone G, Taban C (1992),Immunity, major depression, and panic disorder comorbidity. Biol Psychiatry31:896–908

Ayuso JL, Alfonso S, Rivera A (1989), Childhood separation anxiety and panicdisorder: a comparative study. Prog Neuropsychopharmacol Biol Psychiatry13:665–671

Balon R, Yeragani VK, Pohl R (1989), Higher frequency of separation anxi-ety disorder in panic disorder patients (letter). Am J Psychiatry 146:1351

Bender BG, Annett RD, Ikle D, DuHamel TR, Rand C, Strunk RC (2000),Relationship between disease and psychological adaptation in children inthe Childhood Asthma Management Program and their families: CAMPResearch Group. Arch Pediatr Adolesc Med 154:706–713

Berg I (1976), School phobia in children of agoraphobic women. Br J Psychiatry128:86–89

Biederman J, Faraone SV, Hirshfeld-Becker DR, Friedman D, Robin JA,Rosenbaum JF (2001), Patterns of psychopathology and dysfunction inhigh-risk children of parents with panic disorder and major depression.Am J Psychiatry 158:49–57

Blumenthal MN (1997), Epidemiology and genetics of asthma and allergy.In: Allergy, Kaplan AP, ed. New York: Saunders, pp 407–420

Bussing R, Burket RC (1993), Anxiety and intrafamilial stress in children withhemophilia after the HIV crisis. J Am Acad Child Adolesc Psychiatry32:562–567

Bussing R, Burket RC, Kelleher ET (1996), Prevalence of anxiety disordersin a clinic-based sample of pediatric asthma patients. Psychosomatics37:108–115

Butz AM, Alexander C (1993), Anxiety in children with asthma. J Asthma30:199–209

Capps L, Sigman M, Sena R, Henker B, Whalen C (1996), Fear, anxiety andperceived control in children of agoraphobic parents. J Child PsycholPsychiatry 37:445–452

Coplan JD, Tamir H, Calaprice D et al. (1999), Plasma anti-serotonin andserotonin anti-idiotypic antibodies are elevated in panic disorder.Neuropsychopharmacology 20:386–391

First MB, Spitzer RL, Gibbon M, Williams JBW (1995), Structured ClinicalInterview for DSM-IV Axis I Disorders (SCID). New York: New York StatePsychiatric Institute (available from Michael B. First, NYS PsychiatricInstitute, Biometrics Research, 1051 Riverside Drive, New York, NY 10032)

Garralda ME, Bailey D (1986), Children with psychiatric disorders in primarycare. J Child Psychol Psychiatry 27:611–624

Gittelman R, Mannuzza S, Shenker R, Bonagura N (1985), Hyperactive boysalmost grown up, I: psychiatric status. Arch Gen Psychiatry 42:937–947

Graham PJ, Rutter ML, Yule W, Pless IB (1967), Childhood asthma: a psy-chosomatic disorder? Some epidemiological considerations. Br J Prev SocMed 21:78–85

Hirshfeld DR, Rosenbaum JF, Biederman J et al. (1992), Stable behavioralinhibition and its association with anxiety disorder. J Am Acad Child AdolescPsychiatry 31:103–111

Hollingshead AB, Redlich FC (1958), Social Class and Mental Illness. NewYork: Wiley

Horwitz SM, Morgenstern H, Berkman LF (1985), The use of pediatric med-ical care: a critical review. J Chronic Dis 38:935–945

Kagan J (1994), The physiology of inhibited and uninhibited children. In:Galen’s Prophecy. New York: Westview Press, pp 140–169

SEPARATION ANXIETY, PARENTAL PANIC, ATOPY

J . AM. ACAD. CHILD ADOLESC. PSYCHIATRY, 41 :8 , AUGUST 2002 953

Kentgen LM, Klein RG, Mannuzza S, Davies M (1997), Test-retest reliabil-ity of maternal reports of lifetime mental disorders in their children. JAbnorm Psychol 25:389–398

Klein DF (1964), Delineation of two drug-responsive anxiety syndromes.Psychopharmacologia 5:397–408

Klein DF (1993), False suffocation alarms, spontaneous panics, and relatedconditions: an integrative hypothesis. Arch Gen Psychiatry 50:306–317

Klein RG (1995), Is panic disorder associated with childhood separation anx-iety disorder? Clin Neuropharmacol 18: S7–S14

Klein RG, Mannuzza S (1992), Parent as Respondent Informant Schedule: PARIS.New York: New York State Psychiatric Institute (available from Rachel G.Klein, NYU Child Study Center, 550 First Avenue, New York, NY 10016)

Lemanske RF, Casale TB, Kaliner M (1985), The autonomic nervous systemin allergic disease. In: Allergy, Kaplan A, ed. New York: Churchill Livingstone

Lipsitz JD, Martin LY, Mannuzza S et al. (1994), Childhood separation anx-iety disorder in patients with adult anxiety disorders. Am J Psychiatry151:927–929

Marshall PS (1993), Allergy and depression: a neurochemical threshold modelof the relation between the illnesses. Psychol Bull 113:23–43

McNichol KN, Williams HE, Allan J, McAndrew I (1973), Spectrum of asthmain children, 3: psychological and social components. Br Med J 4:16–20

Merikangas KR, Avenevoli S, Dierker L, Grillon C (1999), Vulnerability fac-tors among children at risk for anxiety disorders. Biol Psychiatry 46:1523–1535

Miller BD, Wood BL (1997), Influence of specific emotional states on auto-nomic reactivity and pulmonary function in asthmatic children. J AmAcad Child Adolesc Psychiatry 36:669–677

Mrazek D, Schuman WB, Klinnert M (1998), Early asthma onset: risk ofemotional and behavioral difficulties. J Child Psychol Psychiatry 39:247–254

Norrish M, Tooley M, Godfrey S (1977), Clinical, physiological, and psy-chological study of asthmatic children attending a hospital clinic. ArchDis Child 52:912–917

Noyes R Jr, Woodman C, Garvey MJ et al. (1992), Generalized anxiety dis-order vs. panic disorder: distinguishing characteristics and patterns ofcomorbidity. J Nerv Ment Dis 180:369–379

Panhuysen CI, Meyers DA, Postma DS, Bleecker ER (1995), The genetics ofasthma and atopy. Allergy 50:863–869

Papp LA, Klein DF, Gorman JM (1993), Carbon dioxide hypersensitivity,hyperventilation, and panic disorder. Am J Psychiatry 150:1149–1157

Papp LA, Martinez JM, Klein DF et al. (1997), Respiratory psychophysiol-ogy of panic disorder: three respiratory challenges in 98 subjects. Am JPsychiatry 154:1557–1565

Perna G, Bertani A, Diaferia G, Arancio C, Bellodi L (1994), Prevalence ofrespiratory diseases in patients with panic and obsessive compulsive dis-orders. Anxiety 1:100–101

Perna G, Bertani A, Politi E, Colombo G, Bellodi L (1997), Asthma and panicattacks. Biol Psychiatry 42:625–630

Pine DS, Klein RG, Coplan JD et al. (2000), Differential carbon dioxide sen-sitivity in childhood anxiety disorders and nonill comparison group. ArchGen Psychiatry 57:960–967

Ramesh C, Yeragani VK, Balon R, Pohl R (1991), A comparative study ofimmune status in panic disorder patients and controls. Acta Psychiatr Scand84:396–397

Rapaport MH (1998), Circulating lymphocyte phenotypic surface markers inanxiety disorder patients and normal volunteers. Biol Psychiatry 43:458–463

Rogeness GA, Cepeda C, Macedo CA, Fischer C, Harris WR (1990), Differencesin heart rate and blood pressure in children with conduct disorder, majordepression and separation anxiety. Psychiatr Res 33:199–206

Schachter EN, Doyle CA, Beck GJ (1984), A prospective study of asthma ina rural community. Chest 85:623–630

Schleyer B, Mannuzza S (1995), Family Informant Schedule and Criteria-Revisedfor DSM-IV (FISC-IV). New York: New York State Psychiatric Institute(available from Salvatore Mannuzza, NYU Child Study Center, 550 FirstAvenue, New York, NY 10016)

Schmidt-Traub S, Bamler KJ (1997), The psychoimmunological associationof panic disorder and allergic reaction. Br J Clin Psychol 36:51–62

Shavitt RG, Gentil V, Mandetta R (1992), The association of panic/agoraphobiaand asthma: contributing factors and clinical implications. Gen HospPsychiatry 14:420–423

Silverglade L, Tosi DJ, Wise PS, D’Costa A (1994), Irrational beliefs and emo-tionality in adolescents with and without bronchial asthma. J Gen Psychol121:199–207

Sly MR (1996), Allergic disorders. In: Textbook of Pediatrics, Behrman RE,Kliegman RM, Arvin AM, eds. Philadelphia: Saunders, pp 610–656

Spinhoven P, Ros M, Westgeest A, Van der Does AJ (1994), The prevalenceof respiratory disorders in panic disorder, major depressive disorder andV-code patients. Behav Res Ther 32:647–649

Spitzer RL, Williams JB, Gibbon M, First MB (1992), The Structured ClinicalInterview for DSM-III-R (SCID). Arch Gen Psychiatry 49:624–629

ten Thoren C, Petermann F (2000), Reviewing asthma and anxiety. RespirMed 94:409–415

Vila G, Nollet-Clemencon C, de Blic J, Mouren-Simeoni MC, ScheinmannP (2000), Prevalence of DSM-IV anxiety and affective disorders in a pedi-atric population of asthmatic children and adolescents. J Affect Disord58:223–231

Wamboldt MZ, Schmitz S, Mrazek D (1998), Genetic association betweenatopy and behavioral symptoms in middle childhood. J Child PsycholPsychiatry 39:1007–1016

Wamboldt MZ, Weintraub P, Krafchick D, Wamboldt FS (1996), Psychiatricfamily history in adolescents with severe asthma. J Am Acad Child Psychiatry35:1042–1049

Weissman MM, Leckman JF, Merikangas KR, Gammon GD, Prusoff BA(1984), Depression and anxiety disorders in parents and children: resultsfrom the Yale Family Study. Arch Gen Psychiatry 41:845–852

Weissman MM, Markowitz JS, Ouellette R, Greenwald S, Kahn JP (1990),Panic disorder and cardiovascular/cerebrovascular problems: results froma community study. Am J Psychiatry 147:1504–1508

Yellowlees PM, Haynes S, Potts N, Ruffin RE (1988), Psychiatric morbidityin patients with life-threatening asthma: initial report of a controlled study.Med J Aust 149:246–249

Yeragani VK, Pohl R, Balon R, Jung I, Kareus R (1991), Heart rate in panicdisorder. Acta Psychiatr Scand 83:79

Zandbergen J, Bright M, Pols H, Fernandez I, de Loof C, Griez EJ (1991),Higher lifetime prevalence of respiratory diseases in panic disorder? Am JPsychiatry 148:1583–1585

Zitrin CM, Ross DC (1988), Early separation anxiety and adult agoraphobia.J Nerv Ment Dis 176:621–625

SLATTERY ET AL.

954 J . AM. ACAD. CHILD ADOLESC. PSYCHIATRY, 41 :8 , AUGUST 2002