Factori de Risc , Erori Cognitive

7/30/2019 Factori de Risc , Erori Cognitive

1/29

http://bmo.sagepub.com

Behavior Modification

DOI: 10.1177/01454455062970162007; 31; 174Behav Modif

Melinda F. CannonCarl F. Weems, Natalie M. Costa, Sarah E. Watts, Leslie K. Taylor and

SymptomsTheir Unique and Specific Associations With Childhood Anxiety

Cognitive Errors, Anxiety Sensitivity, and Anxiety Control Beliefs:

http://bmo.sagepub.com/cgi/content/abstract/31/2/174The online version of this article can be found at:

Published by:

http://www.sagepublications.com

can be found at:Behavior ModificationAdditional services and information for

http://bmo.sagepub.com/cgi/alertsEmail Alerts:

http://bmo.sagepub.com/subscriptionsSubscriptions:

http://www.sagepub.com/journalsReprints.navReprints:

http://www.sagepub.com/journalsPermissions.navPermissions:

http://bmo.sagepub.com/cgi/content/refs/31/2/174SAGE Journals Online and HighWire Press platforms):

(this article cites 44 articles hosted on theCitations

by Amalia Ciuca on October 7, 2008http://bmo.sagepub.comDownloaded from
http://bmo.sagepub.com/cgi/alertshttp://bmo.sagepub.com/cgi/alertshttp://bmo.sagepub.com/subscriptionshttp://bmo.sagepub.com/subscriptionshttp://www.sagepub.com/journalsReprints.navhttp://www.sagepub.com/journalsReprints.navhttp://www.sagepub.com/journalsPermissions.navhttp://www.sagepub.com/journalsPermissions.navhttp://bmo.sagepub.com/cgi/content/refs/31/2/174http://bmo.sagepub.com/cgi/content/refs/31/2/174http://bmo.sagepub.com/http://bmo.sagepub.com/http://bmo.sagepub.com/http://bmo.sagepub.com/http://bmo.sagepub.com/cgi/content/refs/31/2/174http://www.sagepub.com/journalsPermissions.navhttp://www.sagepub.com/journalsReprints.navhttp://bmo.sagepub.com/subscriptionshttp://bmo.sagepub.com/cgi/alerts


2/29

174

Cognitive Errors, AnxietySensitivity, and AnxietyControl Beliefs

Their Unique and Specific

Associations With Childhood

Anxiety Symptoms

Carl F. WeemsNatalie M. CostaSarah E. WattsLeslie K. TaylorMelinda F. CannonUniversity of New Orleans, LA

This study examined the interrelations among negative cognitive errors, anxiety

sensitivity, and anxiety control beliefs and explored their unique and specific

associations with anxiety symptoms in a community sample of youth. Existing

research has suggested that these constructs are related to childhood anxiety dis-

order symptoms; however, additional research is needed to test the interrelations

among negative cognitive errors, anxiety sensitivity, and anxiety control beliefs

and to determine if they show unique and specific associations with anxiety

symptoms. The results of this study indicated that negative cognitive errors, anx-

iety sensitivity, and anxiety control beliefs were associated with each other and

that they demonstrated unique concurrent associations with childhood anxiety

disorder symptoms. Moreover, certain cognitive biases showed specificity in

their association with anxiety symptoms versus depressive symptoms.

Keywords: anxiety; child and adolescents; cognition; anxiety sensitivity;

cognitive errors; control

Anxiety has been broadly conceptualized as a complex response system

involving behavioral, physiological, and cognitive components (e.g.,Barlow, 2002; Lang, 1977). Anxiety disorders are thought to result from

quantitative and/or qualitative deviations in the normative mechanisms of

the anxiety response system (Barlow, 2002; Vasey & Dadds, 2001). Cognitive

models focus on the hypothesis that anxiety disorders can stem from faulty,

Behavior Modification

Volume 31 Number 2

March 2007 174-201

2007 Sage Publications

10.1177/0145445506297016

http://bmo.sagepub.comhosted at

http://online.sagepub.com

http://bmo.sagepub.com/http://bmo.sagepub.com/http://bmo.sagepub.com/http://bmo.sagepub.com/


3/29

Weems et al. / Childhood Anxiety 175

biased, or negative ways of thinking and the biased processing of information

(e.g., Beck, 1976; Ellis, 1962). Although a large number of studies have

shown that cognitive biases are associated with depression in youth and thatvarious forms of biases may help predict depressive symptoms (Garber,

Keiley, & Martin, 2002; see Joiner & Wagner, 1995, for a review), far less

research has been focused on their role in childhood anxiety. Alfano,

Beidel, and Turner (2002) have argued that the literature to date has only

provided modest support for a role of cognitive aberrations (p. 1209) or

cognitive biases in childhood anxiety and anxiety disorders. In particular,

the empirical support for the conceptual distinctions among various cogni-

tive biases and their specificity to anxiety is limited (Weems & Watts, 2005).The purpose of this study was to address some of these gaps in knowledge.

The research on cognitive models of anxiety in youth has focused on biased

interpretation, biased judgment, biased memory, and selective attention (see,

e.g., Vasey & MacLeod, 2001). These various biases are hypothesized to work

together at various stages of information processing to foster and maintain

heightened anxiety. Weems and Watts (2005) developed a model of the cogni-

tive influences on childhood anxiety, which suggests that attention biases may

foster the selective encoding of threat information into memory and that suchselective attention could thus increase the number of negatively biased threat

memories. Memory biases, in turn, may become internalized in cognitive

working models or cognitive schemas, thus fostering interpretive and judg-

ment biases. For example, existing threat memories may bias attention toward

only the threatening part of the situation and away from mitigating aspects of

the situation such as safety signals, thereby fostering anxiety-provoking inter-

pretations. Another way is that existing threat memories may bias the new

interpretation of the event and help to consolidate existing interpretation and

judgment biases. The model thus predicts that the various cognitive biases will

be related to each other but, at least to some extent, will each be incrementally

related to anxiety and anxiety-related distress.

Conceptually, attention and memory are quite distinct and research sug-

gests that they show unique associations with anxiety in youth (Watts &

Weems, in press); however, judgment biases and interpretation biases are very

similar and our review of the literature (see Weems & Watts, 2005) suggests

that additional research and conceptual work is needed to elucidate the

Authors Note: This research was supported in part by a grant from the National Institute of

Mental Health (MH067572) awarded to Carl F. Weems. Correspondence concerning this arti-

cle should be addressed to Carl F. Weems, Department of Psychology, University of New

Orleans, New Orleans, LA 70148; e-mail: [email protected].



4/29

176 Behavior Modification

uniqueness of the concepts and the uniqueness of the associations among

judgment biases, interpretation biases, and anxiety in youth. Interpretive

bias involves being predisposed toward negative or erroneous interpreta-tions of neutral, ambiguous, or potentially threatening stimuli or situations.

Judgment bias involves negative and/or lowered estimates of the individ-

uals coping ability or style. Both interpretive and judgment biases thus

involve a similar cognitive style and so individuals who have a tendency for

negative judgments about their own ability are likely to have negative inter-

pretations of events (and vice versa), and so interpretive and judgment

biases are likely to be related. However, this similarity also raises the pos-

sibility that they are not uniquely associated with anxiety (i.e., variance inone may account for the association between anxiety and the other). The

focus of this study is on two types of interpretive biasesnegative cogni-

tive errors and anxiety sensitivityand a judgment bias, namely, anxiety

control beliefs. These three constructs were chosen because they have been

hypothesized to be important to childhood anxiety problems (e.g., Epkins,

1996; Weems & Silverman, 2006).

Interpretive Biases

As noted, interpretive bias involves being predisposed toward negative

or erroneous interpretations of neutral, ambiguous, or potentially threaten-

ing stimuli or situations. Negatively biased cognitions1 have long been

thought to be core processes in emotional problems such as anxiety disor-

ders (Beck, 1976; Ellis, 1962). Research has shown that clinically anxious

youth presented with ambiguous vignettes and then asked to explain what

was happening in the story are more likely to provide interpretations indi-cating threat than are nonanxious controls (Barrett, Rapee, Dadds, & Ryan,

1996; Chorpita, Albano, & Barlow, 1996).

Negative cognitive errors are thought to be particularly salient to emo-

tional problems in youth (Leitenberg, Leonard, & Carroll-Wilson, 1986)

and concern the interpretation of events and situations in ones life. For

example, catastrophizing involves expecting the worst possible outcome of

an event or situation, overgeneralizing involves believing that a single neg-

ative outcome is representative of all similar future events, personalizinginvolves attributing control over the outcome of negative events to internal

causes, and selective abstraction involves selectively focusing on only the

negative aspects of an event or situation. The existing research on these

cognitive errors using the Childrens Negative Cognitive Error Questionnaire

(CNCEQ; Leitenberg et al., 1986) suggests that they are associated with



5/29

symptoms of anxiety and depression and that certain errors, such as cata-

strophizing and overgeneralizing, may be more related to anxiety, whereas

selective abstraction may be more related to depression (Epkins, 1996;Leitenberg et al., 1986; Leung & Wong, 1998; Weems, Berman, Silverman,

& Saavedra, 2001). However, as theoretically predicted by the cognitive con-

tent specificity hypothesis (Beck, 1976; Beck, Brown, Steer, Eidelson, &

Riskind, 1987; Laurent & Stark, 1993), greater specificity has been obtained

when the content of the cognitions are centered on depressive versus anx-

ious content (Leung & Poon, 2001). Such findings suggest that it is not the

type of bias per se that is specific to anxiety but the content of the bias (see

also Laurent & Stark, 1993).Catastrophic interpretations of anxiety-related sensations have been the

focus of the literature on anxiety sensitivity. The concept of anxiety sensi-

tivity refers to beliefs that anxiety-related sensations have severe and nega-

tive consequences (Reiss, 1991). Defined in this way, anxiety sensitivity

provides specificity to the concept of catastrophizing by focusing on anxiety-

related events and situations. Research indicates that anxiety sensitivity pre-

dicts panic beyond that predicted by trait anxiety in adult samples (e.g.,

Schmidt, Lerew, & Jackson, 1997, 1999), prospectively predicts the devel-opment of panic attacks in youth (Hayward, Killen, Kraemer, & Taylor,

2000; Weems, Hayward, Killen, & Taylor, 2002), and that anxiety sensitiv-

ity is concurrently and prospectively related to self-report of anxiety and

panic symptoms (Ginsberg & Drake, 2002; Lau, Calamari, & Waraczynski,

1996; Silverman, Fleisig, Rabian, & Peterson, 1991). Moreover, studies

have demonstrated that although anxiety sensitivity is related to depression,

it is more uniquely associated with anxiety symptoms in youth (see Dehon,

Weems, Stickle, Costa, & Berman, 2005; Joiner et al., 2002).

Weems et al. (2001) examined the association between negative cogni-

tive errors assessed with the CNCEQ and anxiety symptoms as well as anx-

iety sensitivity assessed with the Childhood Anxiety Sensitivity Index

(Silverman et al., 1991) in a sample of children and adolescents who were

clinic referred for anxiety disorders (N = 251, ages 6-16 years). Resultsindicated that the types of errors, except selective abstraction, were signif-

icantly positively related to self-reported anxiety and anxiety sensitivity

even when controlling for levels of depression (correlations ranged from

.39 to .43). Results of regression analyses indicated that age moderated therelation between the cognitive errors and anxiety. The pattern that emerged

was statistically significant but there were somewhat smaller correlations

between CNCEQ subscale scores and the anxiety measures in children

ages 6 to 11 (average r = ~.35) than adolescents ages 12 to 17 years




6/29

(average r= ~.45). These results suggest that for younger children withanxiety disorders, negative cognitive errors were less strongly related to

their anxious symptoms than for older youth with anxiety disorders.Additional results indicated that selective abstraction was more uniquely

associated with depression and that catastrophizing and overgeneralizing

were more uniquely associated with anxiety symptoms. However, the

unique associations among anxiety, anxiety sensitivity, and CNCEQ scores

were not examined.

Judgment Biases

As noted, judgment bias involves negative and/or lowered estimates of the

individuals coping ability or style. Judgment biases in children can refer to

lowered expectations of their ability to handle threatening situations or

events. Common definitions of control involve a judgment as to ones ability

and have thus been the most common way that judgment biases have been

studied in youth. Barlows (2002) model of anxiety suggests that a perceived

lack of control over external threats (i.e., events, objects, or situations that are

fear producing for an individual) and/or control over negative internal emo-

tional and bodily reactions are central to the experience of anxiety problems;

that is, beliefs that anxiety-related events and sensations are uncontrollable is

part of what makes anxiety a problem for individuals with anxiety disorders.

In other words, nonpathological anxiety in individuals who do not have anx-

iety disorders is differentiated from pathological anxiety by heightened levels

of anxiety in response to the experience of threatening situations but also by

the judgment that they cannot control these events.

Research on control and anxiety has produced an encouraging body ofknowledge. For example, studies suggest that there seems to be a fairly con-

sistent relation between an external locus of control and self-reported anx-

iety in children (Nunn, 1988; Rawson, 1992). Similar findings also have

been evidenced using diverse methods of assessing control (e.g., Capps,

Sigman, Sena, Henker, & Whalen, 1996; Cortez & Bugental, 1995). However,

control is also strongly related to depression as well as anxiety (see Joiner

& Wagner, 1995).

Barlow and colleagues(Barlow, 2002; Rapee, Craske, Brown, & Barlow,1996) model of control provides a more anxiety-specific conceptualization.

Drawing on Barlows (2002) model of control in anxiety disorders, Rapee

et al. (1996) developed a measure of control over anxiety, the Anxiety Control

Questionnaire (ACQ), for use with adults. Ones perception of control over




7/29

anxiety-related events is assessed by the ACQ in terms of control over emo-

tional reactions and frightening events. Specifically, in the ACQ, the internal-

external distinction refers to the stimuli to be controlled (i.e., internal anxietyreactions such as heart palpitations or feelings of panic; external threats

such as a dog or social situations).

Weems, Silverman, Rapee, and Pina (2003) investigated the role of con-

trol beliefs in childhood anxiety disorders using a developmentally modi-

fied version of Rapee et al.s (1996) Anxiety Control Questionnaire (i.e., the

ACQ-C) in a sample of 117 youth ages 9 to17 years. Eighty-six participants

were clinic referred and met diagnostic criteria for an anxiety disorder, and

31 participants were nonreferred comparison participants. Findings indi-cated that perceived control over anxiety-related events was significantly

negatively correlated with youth self-reported anxiety symptoms and that

youth with anxiety disorders reported significantly lower perceived control

about anxiety than the nonreferred youth. Results of logistic regression

analysis indicated that the perceptions of control over anxiety predicted

anxiety disorder status even when controlling for anxiety levels using an

existing measure of anxiety as well as locus of control. Although this initial

research is encouraging, the specificity of anxiety control beliefs to anxietyversus depression and its association to other forms of cognitive bias has

not been examined.

Summary

There is evidence that negative cognitive errors, anxiety sensitivity, and

anxiety control beliefs are associated with anxiety in youth. However,

research is needed to examine the interrelations among negative cognitiveerrors, anxiety sensitivity, and anxiety control beliefs and to examine if they

are each uniquely associated with anxiety. For example, anxiety sensitivitys

association with anxiety may be explained by the more general cognitive

bias of catastrophizing. Similarly, it could be argued that personalizing or

attributing control over the outcome of negative events to internal causes is

tapping the same cognitive process as control over anxiety. Moreover, the

judgment that anxiety sensations are uncontrollable might be tapping the

same beliefs that are central to anxiety sensitivity (i.e., they are not empiri-cally unique). Theoretically, biases such as control over anxiety involve a

perceived lack of ability, competence, or skill and are thus centered on the

individual making the judgment, whereas interpretive biases involve dispro-

portionately negative interpretations of stimuli or situations and are thus




8/29

centered on a context, stimuli, or external event (i.e., they theoretically should

be empirically unique predictors of anxiety). According to the cognitive con-

tent specificity hypothesis (Beck et al., 1987), the concepts of anxiety sensi-tivity and anxiety control should show greater specificity to anxiety than

depression. Without research that tests the uniqueness and specificity of

these constructs (e.g., Do measures of personalizing, catastrophizing, anxi-

ety sensitivity, and control predict unique variance in anxiety symptoms and

how specific are they to anxiety symptoms vs. depression?), the conceptual

validity of the constructs remains questionable.

The Present Study

The purpose of this study was to extend existing research by examining

the interrelations among negative cognitive errors, anxiety sensitivity, and

anxiety control beliefs and by examining their unique associations with

anxiety symptoms in a community sample of youth. Based on the model

proposed by Weems and Watts (2005), theoretical basis for the constructs,

and existing research, we predicted that catastrophizing, anxiety sensitivity,

and anxiety control would each predict unique variance in anxiety disordersymptoms. We also hypothesized that anxiety sensitivity and anxiety con-

trol would show specific associations with anxiety (i.e., would be related to

anxiety even when controlling for depression). Based on previous research,

we predicted that selective abstraction would show unique prediction of

symptoms of depression (e.g., Epkins, 1996; Weems et al., 2001). We also

examined the 1-year test-retest reliability of the measures of negative cog-

nitive errors (CNCEQ), anxiety sensitivity (Childhood Anxiety Sensitivity

Index [CASI]), and anxiety control beliefs (ACQ-C) and examined if thesepredicted anxiety symptoms after 1 year in a subsample (n = 52) of thechild participants.

Given the role that age may play in the associations, we recruited a sam-

ple of youth from a wide age range. This was done so that the potential

moderating role of age could be examined. Based on previous findings in

the depression literature (e.g., Nolen-Hoeksema, Girgus, & Seligman,

1992; Turner & Cole, 1994) and previous research (Calamari et al., 2001;

Rabian, Embry, & MacIntyre, 1999; Weems et al., 2001; Weems, Hammond-Laurence, Silverman, & Ginsburg, 1998), we expected some moderation of

CNCEQ subscales by age, but not anxiety sensitivity. In the absence of any

data on the ACQ-C, we did not make a priori predictions. By using the term

some moderation, we expected that the association would be weaker but




9/29


that the measures of cognitive biases would still be significantly associated

with anxiety in younger children. However, we expected that age-related

differences in the size of the association between cognitive biases and anx-iety may be specific to clinical samples because of overall larger effect sizes

in such samples. The potential moderating role of gender and ethnicity also

was tested; however, based on Weems et al. (2001), we expected that gender

would not be a moderator. The moderating role of ethnicity was examined

for exploratory purposes in response to the call for greater attention to eth-

nic differences in childhood anxiety research (Cooley & Boyce, 2004).

In addition, although youth are probably in the best position to report on

their cognitions, reliance solely on self-report of anxiety symptoms in test-ing the validity of these measures may have biased estimates of the associ-

ation. However, there tends to be fairly low correspondence between parent

and child reports of the childs behavior (e.g., Achenbach, McConaughy, &

Howell, 1987, in a review of the literature, report an average correlation of

r= .25). Such discrepancies among reporters can lead to different or incon-sistent conclusions about the prediction of behavioral problems (De Los

Reyes & Kazdin, 2004). Reasons for the bias in reporting anxiety symp-

toms may stem from the parent or from the child. For example, bias may bedue to the child underreporting his or her own anxiety for reasons of social

desirability, or parents may underreport relative to the child because the

nature of anxiety makes it less salient to parents and thus they are not aware

of true levels of anxiety in their child. De Los Reyes and Kazdin (2004)

recently examined different options for examining informant discrepancies.

They concluded that the standardized difference between reporters is a pre-

ferred measure of discrepancy between parent and child reports because it

correlates equally with each of the sources from which it is derived.

Drawing from this recommendation, we employed a technique for handling

discrepancy when predicting parent-reported symptoms with child-reported

cognitive biases. Results obtained from parents alone and from combining

parent and child reports of the childs anxiety also are presented.

Method

ParticipantsData were collected from a socioeconomic and ethnically diverse sample

of 145 youth (ages 6-17 years, Mage = 11.36 years, SD = 3.5 years)2 andtheir primary caregiver (mothers = 90%, fathers = 7%, grandparents = 3%).



10/29

Participants were Euro-American (42%), African American (42%), Hispanic

(8%), Asian (2%), and 6% were of other or mixed ethnic backgrounds, with

a median family income of between $20,000 and $40,000 a year. Fifty-fivepercent of the sample was female. Families were recruited through the adult

students enrolled in courses at the University of New Orleans (UNO) as

well as through area schools and media outreach. Participants received a

small monetary reward as compensation for participating in the research

study (paid to the parent for the child).

Children were excluded if parents indicated during an initial assessment

screening that the child had a history of one or more of the following

diagnosesall pervasive developmental disorders, mental retardation, selec-tive mutism, organic mental disorders, schizophrenia, and other psychotic

disordersor were at risk for harm to self or others (only one child was

excluded whose parent indicated the child had a diagnosis of pervasive

developmental disordernot otherwise specified [PDD-NOS]). Interested

families were informed that we were conducting a study of youth behav-

iors, emotions, and anxiety and that they could receive a free screening for

anxiety-related problems. However, potential participants were told that

families are eligible to participate regardless of whether they have anxietyproblems. This recruitment process was designed to help normalize the dis-

tribution of anxiety symptoms to facilitate the use of parametric statistics.

However, only 6% of children or mothers were currently on any form of

psychotropic medications. On the basis of our assessment, 9 children (6%)

were referred to further assessment or intervention services. All data used

in this study were collected before families were referred to or enrolled in

intervention services.

A representative subsample (n = 52)3 of the child participants completedthe measures again approximately 1 year (11-14 months) after the initial

assessment to obtain 1-year test-retest reliability estimates and to examine

the predictive ability of negative cognitive errors, anxiety sensitivity, and

anxiety control beliefs. The mean age of the follow-up sample at Time 1

was 11.15 years (SD = 3.2 years). Participants were Euro-American (42%),African American (46%), Hispanic (7%), and 4% were of other or mixed

ethnic backgrounds, with a median family income of between $20,000 and

$40,000 a year. Sixty-one percent of the sample was female. Completers of

this Time 2 follow-up were compared to the rest of the sample on Time 1variables. The follow-up sample did not differ on age, gender, overall eth-

nic distribution, income, anxiety, and depression scores or any of the cog-

nitive variables.




11/29

Measures

The CNCEQ (Leitenberg et al., 1986) was used to assess catastrophiz-ing, overgeneralization, personalizing, and selective abstraction. This

24-item measure is designed to assess the four errors via four theoretically

derived subscales. Each subscale contains six questions. Each item on each

subscale presents the child with a hypothetical vignette and a negative inter-

pretation of the vignette to which the child responds if he or she would

interpret the situation in a similar fashion. For example, one of the items

assessing selective abstraction has the hypothetical situation of playing bas-

ketball where it is indicated that during the game the child scored five bas-

kets and missed two easy shots. After the game, the child thought, I played

poorly. Children are asked to rate on a 5-point scale how similar the

thought was to their own thoughts in a similar situation (e.g., 1 = not at alllike I would think, 5 = almost exactly like I would think). The CNCEQ hasdemonstrated acceptable internal consistency, test-retest reliability, and

construct validity estimates (Leitenberg et al., 1986; Weems et al., 2001).

For example, in terms of convergent validity, Mazur, Wolchik, and Sandler

(1992) found that total distortion scores were significantly related to

childrens anxiety scores as measured by the Revised Childrens ManifestAnxiety Scale (RCMAS) while controlling for gender using partial corre-

lations (partial r= .28).The CASI (Silverman et al., 1991) was used to assess anxiety sensitiv-

ity. The CASI is an 18-item measure designed to assess childrens fear of

different symptoms of anxiety. Children rate each question by selecting one

of three choices (none, some, or a lot). Each item is scored with a 1, 2, or

3. Example questions are, It scares me when I feel shaky and It scares

me when I feel nervous. The CASI has been shown to have satisfactoryreliability estimates (Silverman et al., 1991). For example, Cronbachs

alpha coefficient has been estimated at .87 and 2-week test-retest reliability

at .76. With regard to validity, the CASI has been shown to (a) concurrently

relate to panic symptoms, fears, and negative cognitive errors; (b) predict

anxiety-relevant responding to behavioral-stress challenge tasks; and (c) be

prospectively associated with panic symptoms (e.g., Ginsburg & Drake,

2002; Lau et al., 1996; Rabian et al., 1999; Silverman et al., 1991; Weems

et al., 2001).The ACQ-C (Weems et al., 2003) was developed by adapting the method

used in Rapee et al. (1996). Questions were designed to measure perceived

lack of control over external threats (e.g., events, objects, or situations that

are fear producing for an individual) and control over negative, internal,




12/29

emotional, and bodily reactions associated with anxiety (e.g., shaking or

trembling, the subjective experience of anxiety). In adapting the methods

used by Rapee et al. (1996), Rapee et al.s items were subjected to a devel-opmental analysis of the wording by the authors and modified to render

them understandable by children. For example, phrasing and wording on

the question, There is little I can do to influence peoples judgments of

me was changed to I can change the way that people feel about me.

Children were asked to rate their agreement with each question along a

rating scale as follows: 0 (none), 1 (a little), 2 (some), 3 (a lot), or 4 (very, very

much). Total control belief score is obtained by summing the items. Internal

consistency estimates (i.e., coefficient alpha) for the total scale were .94and .92 in two independent samples (Weems et al., 2003). The ACQ-C also

has produced good validity estimates. For example, the ACQ-C had con-

vergent validity with the NSLOC (r= .22) and RCMAS (r= .47).The Revised Child Anxiety and Depression Scales (RCADS; Chorpita,

Yim, Moffitt, Umemoto, & Francis, 2000; Spence, 1997) were used to

assess symptoms of anxiety disorders and depression. The RCADS is a

47-item instrument that assesses symptoms of each anxiety disorder (except

posttraumatic stress disorder [PTSD] and specific phobias) and depressionbased on the Diagnostic and Statistical Manual of Mental DisordersIV

(DSM-IV) criteria (American Psychiatric Association, 1994). The scale is

scored 1 (never), 2 (sometimes), 3 (often), and 4 (always). The RCADS is

an adaptation of the Spence Anxiety Scales (Spence, 1997). Chorpita et al.

(2000) modified the Spence scales forDSM-IVand evaluated the RCADS

by examining the measures factorial validity in a school sample of 1,641

children and adolescents (ages 6.2-18.9 years) and its reliability and validity

in an independent sample of 246 children and adolescents (ages 8.3-18.3 yrs).

The results suggested an item set and factor definitions that were consistent

with DSM-IV anxiety disorders and depression. Moreover, the RCADS

demonstrated convergent validity with existing measures of childhood anx-

iety and depression (Chorpita et al., 2000).

Parents also completed a parent version of the RCADS (RCADS-P)

designed identical to the RCADS with minor modifications (i.e., wording

was changed from I to My child). Because the RCADS-P has not been

used before, the Revised Childrens Manifest Anxiety Scale (RCMAS-P)

was completed by parents (Pina, Silverman, Saavedra, & Weems, 2001;Reynolds & Richmond, 1978) and was administered to assess convergent

validity with the RCADS-P. The RCMAS-P has good validity estimates

(Pina et al., 2001) and asks parents to rate their childs anxiety with virtu-

ally the same items as the RCMAS (i.e., wording is changed from I to




13/29


My child), and thus, 28 items are summed from yes or no responses to

yield a Total Anxiety score (nine items comprise a lie scale). Each item is

scored with a 0 or 1. Convergent validity of the RCADS-P anxiety scalewith the RCMAS-P was .81 in the full sample, .78 in the youth ages 6 to 11

years, and .84 in the youth ages 12 to 17 years (allps < .01). Cross-informantvalidity of the RCADS-P anxiety scale with the RCADS anxiety was .27 in

the full sample, .25 in the youth ages 6 to 11 years, and .27 in the youth ages

12 to 17 years (allps < .05). RCADS-P depression scale was significantlycorrelated with RCMAS-P scores: .73 in the full sample, .69 in the youth

ages 6 to 11 years, and .80 in the youth ages 12 to 17 years (allps < .01).

Procedures

Informed consent was obtained from the parent and informed assent was

obtained from the child. Completion of the assessment took place in a quiet

room and the child completed the assessment in a separate room from the

parent. Both the youth and parent were greeted and given a general

overview of the assessment procedures. Standardized specific instructions

were then given to the parent and child separately. Youth completed themeasures and were assisted as necessary by trained research assistants (e.g.,

young participants were read the assessment battery by research assistants

who monitored the childs comprehension of the questions). At the conclu-

sion of the study, participants were debriefed and given a small monetary

reward. Missing or incomplete data on one or more measures was handled

by pair- or analysis-wise (when more than two variables) deletion of miss-

ing cases (Tabachnick & Fidell, 2001). Age groups4 (children, ages 6-11

years, and adolescents, ages 12-17) were formed on the basis of sample

size, predicted change in cognitive development (Piaget, 1950, 1983), and

for consistency with past research for dichotomous age analyses.

Results

Preliminary Analyses

Means and standard deviations for each of the measures for the total

sample and by age and gender are presented in Table 1. Examination of thecognitive and symptom scores ranges and skew indicated acceptable levels

for the planned analyses. Results of 2 (gender) 2 (age group) ANOVAsare summarized in Table 1 and indicated that boys and girls differed on



14/29


RCADS anxiety scores, CASI scores, and ACQ-C scores and that the age

groups differed on RCADS anxiety and CASI scores. Because the sample

sizes for the other ethnicities were small, the role of ethnicity was consid-

ered separately and only the African American and Euro-American partici-

pants were used for the ethnic comparisons. Results ofttests indicated that

African American and Euro-American participants significantly differed on

CNCEQ catastrophizing (African American M = 13.16, SD = 5.5; Euro-AmericanM= 11.01, SD = 4.0), t(120) = 2.46,p < .05; CNCEQ personal-izing (African AmericanM= 14.69, SD = 5.4; Euro-AmericanM= 12.43,SD = 4.4), t(120) = 2.44, p < .05; CASI (African American M = 31.77,SD = 6.2; Euro-American M= 27.34, SD = 6.8), t(120) = 3.76, p < .001;and RCADS anxiety (African American M = 69.72, SD =17.9; Euro-

AmericanM=59.39, SD =14.7), t(120) = 3.48,p < .01.Pearsons correlations among the measures and internal consistency are

presented in Table 2 and indicated that child-reported RCADS anxiety

scores were significantly but modestly correlated with each of the cognitive

measures. RCADS anxiety was significantly and highly correlated with

Table 1

Means and Standard Deviations for Cognitive and Symptom

Measures by Age and Gender

Full Sample 6-11 Years 12-17 Years Boys Girls

Measure M(SD) M(SD) M(SD) M(SD) M(SD)

1. RCADS-(A)a,b 64.15 (16.7) 68.90 (17.6) 58.83 (14.0) 59.83 (16.1) 67.79 (16.5)

2. RCADS-(PA) 51.55 (10.3) 52.48 (10.0) 50.33 (10.8) 51.51 (10.6) 51.58 (10.2)

3. CASIa,b 29.31 (6.9) 31.25 (6.8) 27.19 (6.4) 26.89 (6.3) 31.30 (6.8)

4. ACQ-Cb 71.03 (20.1) 70.51 (19.2) 71.60 (21.1) 76.82 (22.1) 66.02 (16.8)

5. CNCEQ-CT 12.09 (4.8) 12.01 (4.8) 12.17 (4.8) 12.17 (5.3) 12.01 (4.4)

6. CNCEQ-OG 12.45 (5.0) 12.90 (5.2) 11.95 (4.8) 11.92 (4.9) 12.90 (5.1)7. CNCEQ-PS 13.22 (5.1) 13.98 (5.3) 12.38 (4.7) 13.39 (5.1) 13.06 (5.0)

8. CNCEQ-SA 12.38 (4.6) 12.59 (5.2) 12.15 (4.0) 12.00 (4.3) 12.71 (5.0)

9. RCADS-(D) 16.71 (4.8) 17.39 (4.9) 15.93 (4.7) 15.98 (4.6) 17.31 (4.8)

10. RCADS-(PD) 13.41 (3.5) 13.22 (3.5) 13.60 (3.7) 13.48 (4.1) 13.34 (3.0)

Note: RCADS = The Revised Child Anxiety and Depression Scales; A = Anxiety; D =

Depression; PA = Parent-Completed Anxiety; PD = Parent-Completed Depression; CASI =

Childhood Anxiety Sensitivity Index; ACQ-C = Anxiety Control Questionnaire for Children;

CNCEQ = Childrens Negative Cognitive Error Questionnaire; CT = Catastrophizing; OG =

Overgeneralizing; PS = Personalizing; SA = Selective Abstraction.a. Significant age group difference.

b. Significant gender difference.



15/29

Table2

Internal

ConsistencyandCorre

lationsAmongtheMea

suresofCognitiveBias

,Anxiety,andDepression

1

2

3

4

5

6

7

8

9

10

1.RCADS-(A)

2.RCADS-(PA

)

.27**

3.CASI

.60**

.07

4.ACQ-C

.37**

.12

.23**

5.CNCEQ-CT

.31**

.04

.24**

.10

6.CNCEQ-OG

.41**

.01

.27**

.24**

.67**

7.CNCEQ-PS

.33**

.09

.32**

.09

.66**

.61**

8.CNCEQ-SA

.35**

.10

.25**

.21*

.66**

.64**

.

64**

9.RCADS-(D)

.73**

.21*

.43**

.29**

.35**

.39**

.

31**

.40**

10.RCADS-(PD

)

.11

.74**

.06

.08

.03

.00

.

09

.07

.19*

Coefficientalpha

.93

.93

.86

.91

.73

.74

.

74

.71

.71

.82

Age6-11(n=75)

.92

.91

.83

.89

.71

.73

.

71

.73

.65

.83

Age12-17(n=

70)

.93

.95

.88

.94

.76

.76

.

77

.67

.77

.82

Note:RCADS=

TheRevisedChildAnxietyandDepressionScales;A

=

Anx

iety;D

=

Depression;PA

=Parent-CompletedAnxiety;PD

=

Parent-

CompletedDepr

ession;CASI=

ChildhoodAnxietySensitivityIndex;ACQ-C

=

AnxietyControlQuestionnaireforChildren;CNCEQ

=C

hildrens

NegativeCognitiveErrorQuestionnaire;CT=

Catastrophizing;OG

=

Overgeneralizing;PS=

Personalizing;

SA

=

SelectiveAbstraction.

*p

Factori de Risc , Erori Cognitive

Documents

Transcript of Factori de Risc , Erori Cognitive