Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2016 Aug 16.
Published in final edited form as: J Affect Disord. 2014 Nov 21;174:180–187. doi: 10.1016/j.jad.2014.11.024

Anxiety Comorbidity in Bipolar Spectrum Disorders: The Mediational Role of Perfectionism in Prospective Depressive Symptoms

Jared K O’Garro-Moore 1, Ashleigh Molz Adams 2, Lyn Y Abramson 3, Lauren B Alloy 4
PMCID: PMC4986825  NIHMSID: NIHMS809746  PMID: 25499686

Abstract

Background

Bipolar spectrum disorders (BSDs) are highly comorbid with anxiety, which is associated with an extended duration and exacerbation of depressive symptoms. Unfortunately, the underlying mechanisms are not known. This study examined the role of maladaptive cognitive styles in the co-occurrence of BSDs and anxiety disorders and prediction of depressive symptoms.

Methods

Participants included 141 young adults (69.6% female, mean age= 20.24, SD= 2.11), in one of three groups: a BSD group (bipolar II, cyclothymia, n=48), a comorbid BSD/Anxiety group (n=50), and a demographically-matched healthy control group (n=43), who were followed prospectively. Participants completed the Cognitive Style Questionnaire (CSQ), Depressive Experiences Questionnaire (DEQ), Dysfunctional Attitudes Scale (DAS), Sociotropy Autonomy Scale (SAS), Halberstadt Mania Inventory (HMI) and Beck Depression Inventory (BDI) at the initial assessment. One year later, participants completed the BDI and HMI again to assess severity of depressive and hypomanic/manic symptoms.

Results

A Multivariate Analysis of Co-Variance (MANCOVA) revealed significant differences between the three groups on their DAS Perfectionism, DEQ Dependency, DEQ Self-Criticism, CSQ Negative, SAS Autonomy, and Time 2 BDI scores, with the BSD/Anxiety group scoring higher than the BSD only group on all measures except the CSQ. Preacher and Hayes' (2008) bootstrapping method was used to test for mediational effects of the significant cognitive style measures on depressive symptoms at follow-up. The 95% confidence intervals for the indirect effect of group on follow-up depressive symptoms through DAS Perfectionism did not include zero, indicating the presence of a significant mediating relationship for perfectionism.

Limitations

This study only used two waves of data; three waves of data would allow one to investigate the full causal effect of one variable on another. Further, a comorbid anxiety diagnosis consisted of any anxiety disorder. Further research should separate groups by their specific anxiety diagnoses; this could afford a more complete understanding of the effect of types of anxiety on prospective depressive symptoms.

Conclusions

After taking into account initial levels of depressive and hypomanic/manic symptoms, we found that those with BSD/anxiety comorbidity experienced more severe depressive symptoms, but not more severe hypomanic/manic symptoms. Further, their more severe prospective depressive symptoms are explained by a perfectionistic cognitive style.

Affective lability is a common feature of human emotion. When this fluctuation in mood becomes extreme, it is characteristic of a group of disorders known as bipolar spectrum disorders (BSDs). BSDs can range from the relatively mild, such as cyclothymia, which involves fluctuations between hypomanic and subthreshold depressive symptoms, to the most severe, bipolar I disorder (BD I), characterized by extreme highs (e.g., mania) and lows (e.g., major depression) of mood and behavior (Miklowitz & Johnson, 2006; Parker, McCraw & Fletcher, 2012). BSDs affect approximately 4.4% of the United States population (Merikangas et al., 2007) and are associated with many deleterious physical and interpersonal outcomes (Goodwin & Jamison, 2007). Suicidal ideation is common in BSDs (Simon et al, 2007), and 10–15% of individuals with BSD successfully carry out suicide attempts (Harrison & Barraclough, 1997; Angst et al., 2005). In addition, individuals with BSDs exhibit impairment in functioning, including high rates of divorce (Goodwin & Jamison, 2007) and substance abuse (Angst et al., 2002; Alloy et al., 2009b).

Mood disorders including BSDs commonly are comorbid with anxiety disorders in both clinical and epidemiological samples (Akiskal et al., 2006; McElroy et al. 2001; Zimmerman et al. 2002; Kessler et al 2003; Goes et al. 2012). Approximately 74.9% of individuals with BD will meet diagnostic criteria for an anxiety disorder during their lifetime (Merikangas et al, 2007; Provencher, Guimond, & Hawke, 2012). Among those BD individuals who do meet diagnostic criteria for an anxiety disorder, social anxiety disorder, specific phobia, and generalized anxiety disorder (GAD) have the highest comorbidity rate with estimates of 37.8%, 35.5%, and 29.6%, respectively (Merikangas et al., 2007). The comorbidity between bipolar and anxiety disorders is associated with an exacerbation and extended duration of depressive symptoms (Simon et al., 2007; Goes et al., 2012; Coryell et al., 2012). Specifically, comorbidity with obsessive compulsive disorder (OCD), panic disorder (PD), and social phobia have been associated with more frequent suicidal ideation (Simon et al., 2007) and suicide attempts (Perroud et al., 2007; Saunders et al., 2012). Along with a more severe depressive course, comorbid individuals have been shown to display shorter euthymic periods (Simon et al., 2004; Saunders et al., 2012) and a longer duration of time to remission of affective episodes (Simon et al., 2004; Sala et al., 2012). Moreover, bipolar disorder individuals with comorbid panic-agoraphobia diagnoses have been found to experience a higher number of hypomanic episodes (Akiskal et al., 2006). Regarding the treatment of BSD with anxiety comorbidity, these individuals exhibit a poorer response to mood-stabilizing medications and greater risk of medication-induced mania (Henry et al., 2003; Pini et al., 2003; Sala et al., 2012). Given that those individuals who experience BSD/anxiety comorbidity display worse symptom presentation and outcome than those with a BSD alone, one explanation may be that dysfunctional cognitive styles mediate this association.

Unforturnately, there is a dearth of knowledge as to the underlying mechanisms that may explain comorbidity between BSDs and anxiety. Inasmuch as they do co-occur, it is not unreasonable to postulate that they share common risk factors, specifically maladaptive cognitive styles. Cognitive or personality styles that have been associated with a vulnerability to both anxiety and BSDs include high levels of sociotropy and autonomy, dependency and self-criticism, as well as perfectionistic attitudes and negative inferential styles (Bögels and Zigterman, 2000; Alloy et al. 2009a; Alloy et al., 2012).

Sociotropy and autonomy describe personality characteristics or cognitive styles that, when extreme and dysfunctional, are vulnerability markers for depression (Beck, 1983). Sociotropy, which can also be thought of as interpersonal dependency, has been shown to be high in depressed individuals (Hirschfeld et al, 1983; Clark and Beck, 1991). Beck (1983) argued that a highly sociotropic individual not only aims to maintain close relationships, but also seeks the approval and acceptance of others. This type of individual is likely to exhibit depression when they are confronted with a deterioration of interpersonal relationships (e.g., divorce, moving away from a friend, etc). Conversely, an individual high in autonomy is one who invests in preserving independence, prefers solitary activities, and values work and achievement. Accordingly, failures in achievement or goal-directed behaviors (e.g., demotion, failure on an exam, etc.) are likely to lead to depression for highly autonomous individuals (Beck, 1983; Clark and Beck, 1991). Recent correlational studies have shown a link between sociotropy/autonomy and unipolar depression and anxiety (e.g., Sutton et al., 2011). Brenning and colleagues (2013) found that high levels of initial autonomy predicted depression prospectively. Similarly, high levels of sociotropic personality style predicted significant increases in symptoms of depression and anxiety (Alford and Gerrity, 1995; Masih et al, 2007). Brown et al. (1998) found that individuals high on autonomy scored higher on measures of social anxiety and general anxiety. In addition, autonomy is greater in BSD individuals than healthy controls and predicts onset of hypomanic/manic episodes (Alloy et al., 2009a). Thus, sociotropy and autonomy may help to explain BSD/anxiety comorbidity.

Another two negative cognitive styles similar to sociotropy and autonomy that, when dysfunctional, have been linked to high levels of depression are dependency and self-criticism (Blatt et al, 1976). Dependency can be thought of as a coping mechanism that is triggered to offset feelings of worthlessness and guilt, or as a function of needing support or feeling weak (Blatt et al., 1976). This construct manifests in approval- or affection-seeking behavior. Alternatively, self-critical individuals value their achievements, independence, and freedom from control by others, but measure their achievements based on very high internalized standards (Blatt, 1974; Blatt et al., 1976). However, their unreasonably high standards set them up for feelings of failure, disappointment, guilt, and self-blame.

It is worth noting that sociotropy and dependency and autonomy and self-criticism seem to be related constructs. Recent research has attempted to understand whether they represent complementary dimensions of vulnerability. Shahar et al. (2008) demonstrated with confirmatory factor analysis that self-criticism, sociotropy, and dependency can be considered three dimensions that have a significant, positive association with depression. Simply put, these factors are the seminal cognitive styles that may predispose an individual to depression. Interestingly, these findings also have been replicated with symptoms of anxiety. Cohen et al. (2013) found that self-criticism predicted depressive symptoms, whereas dependency predicted social anxiety symptoms. Moreover, Alloy et al. (2009a) found that self-criticism is higher in individuals with BSD than healthy controls and that higher self-criticism also predicts onsets of hypomanic/manic episodes in BSD individuals.

In Beck’s (1967, 1987) cognitive theory of depression, negative attitudes that an individual has towards oneself (e.g., “I’m stupid”), the outside world (e.g., “Everyone knows I’m stupid”), and one’s future (e.g., “I’ll always be stupid”) are not only characteristics of depressed people, but also provide vulnerability to depression. These dysfunctional attitudes can be further described as perfectionistic, self-evaluative responses and beliefs that follow a negative life event in an if-then manner (e.g., “If someone I love doesn’t love me, then I’m worthless”). Similarly, according to the Hopelessness Theory of depression (Abramson, Metalsky, & Alloy, 1989), negative inferential styles, which are comprised of the tendency to attribute stressful life events to stable and global causes, infer that negative consequences will follow from negative events, and infer that one is deficient or unworthy, also provide vulnerability to depression.

Longitudinal studies have shown that dysfunctional attitudes and negative inferential styles predict increases in depressive symptoms and first onset of major depressive episodes in unipolar samples, as well as depressive and hypomanic symptoms in bipolar spectrum samples (Alloy et al, 1999; 2006; Francis-Raniere, Alloy, & Abramson, 2006; Reilly-Harrington, Alloy, Fresco, & Whitehouse, 1999; Fresco et al., 2001; Alloy et al., 2012). In addition, dysfunctional attitudes and negative attributional styles have been found in anxious individuals (e.g., Heimberg et al., 1989; Sanz and Avia, 1994; Burns & Spangler, 2001). However, the extent to which dysfunctional attitudes and negative inferential styles are associated with comorbidity between anxiety and BSDs is not known.

The present study builds upon existing literature and evaluates the role of maladaptive cognitive styles (sociotropy/autonomy, dependency/self-criticism, dysfunctional attitudes, negative inferential style) in the comorbidity of anxiety and BSD. We hypothesized that: (i) individuals with comorbid BSD and anxiety disorders would have more maladaptive cognitive styles than individuals with BSD only and healthy controls, (ii) individuals with comorbid BSD and anxiety disorders would exhibit more severe depressive and hypomanic/manic symptoms prospectively than those with BSD alone, and (iii) the prospective association between comorbid BSD and anxiety and more severe depressive and hypomanic/manic symptoms would be mediated by maladaptive cognitive styles.

Methods

Participants and Procedures

Participants in this study were drawn from the Temple University site of the Longitudinal Investigation of Bipolar Spectrum Disorders (LIBS) Project. This project examined various risk factors that may influence the course of BSDs. A two-phase screening process was used to select the appropriate participants. In Phase I, approximately 7,500 individuals completed the General Behavior Inventory (GBI; Depue, Krauss, Spoont, & Arbisi, 1989). Based on the GBI cutoff score criteria (see measures), 12.3% of the sample was potentially eligible for the bipolar spectrum disorder (BSD) group, and 65% was potentially eligible for the healthy control group. Approximately 2 weeks to 4 months after Phase I, 684 potentially eligible individuals completed a lifetime diagnostic interview for Phase II screening.

In Phase II, inclusion criteria for the BSD group required a current or lifetime diagnosis of bipolar II disorder, cyclothymia, or bipolar disorder not otherwise specified (NOS), as determined by the Diagnostic and Statistical Manual of Mental Disorder, 4th Edition Text Revision (DSM-IV-TR; American Psychiatric Association, 2000) or Research Diagnostic Criteria (RDC; Spitzer, Endicott & Robins, 1978). The bipolar NOS diagnosis was given to individuals who had experienced one of three types of symptomotology: (a) hypomanic episode(s) but no diagnosable depressive episodes, (b) a cyclothymic mood pattern with periods of affective disturbance that did not meet frequency or duration criteria for hypomanic and depressive episodes, or (c) hypomanic and depressive episodes not meeting frequency criteria for a diagnosis of cyclothymia.

Participants were excluded from the BSD group if they had experienced a manic episode as defined by the DSM-IV-TR or RDC criteria, as this would indicate a bipolar I diagnosis. Participants with bipolar I disorder were excluded because a major goal of the LIBS project was to predict subsequent conversion to bipolar I (onset of manic episode). The control group had two exclusion criteria: (a) current or past diagnosis of any Axis I disorder, and (b) family history of a mood disorder. Of the 384 potential BSD group participants from Phase I, 137 (35.7%) met criteria for a BSD (47 with cyclothymia or bipolar NOS and 90 with bipolar II), and 127 of 300 potential control participants from Phase I (40.2%) met criteria for the healthy control group. The final sample consisted of a total of 220 participants, with BSD group participants matched with respect to gender, age, and ethnicity by the control group.

Of the 220 participants that qualified following the first two screening phases, 79 were excluded from the present analyses because they were missing some of the data needed for the present study. These 79 participants did not significantly differ from the remaining 141 participants on the basis of gender (χ2[2]= .68, p = .71), age (t[2]=.06. p = .95, ethnicity (χ2[2]= 2.78, p = .60, or bipolar spectrum diagnosis (χ2[2]= 2.35, p = .31). Thus, the final sample for this study included 98 in the BSD group (65.33% female) and 43 in the control group (68.18% female). Of the 98 BSD participants, 36% met criteria for cyclothymia or bipolar NOS (51.85% female) and 64% met criteria for bipolar II disorder (75% female). Additionally, for the present study’s aims, 55% of the BSD group was allocated to a Bipolar/Anxiety (BSD/Anx) group on the basis of a comorbid lifetime anxiety diagnosis. Specific comorbid anxiety diagnoses in the BSD/Anx group were DSM GAD 15.4%, RDC GAD 21.5%, DSM PD 6.2%, RDC PD 4.6%, DSM PTSD 30.8%, DSM OCD 18.5%, and DSM Anx NOS 3.1%

Thus, the final sample for this study included 48 participants in the BSD group, 50 in the BSD/Anx group, and 43 in the Control group. In the BSD group, 39% of the participants met RDC criteria for BDII, whereas 17% met DSM-IV-TR criteria. In the BSD/Anx group, 76.9% of the participants met RDC criteria for BDII, and 53.8% met DSM-IV-TR criteria. Prior to entering the study, all participants provided informed written consent. Consistent with the longitudinal nature of the study, all participants were asked to complete various measures of life events and symptoms every four months. Data included in this study were compiled from the first assessment, Time 1 (T1), and the one-year follow-up, Time 2 (T2).

Measures

General Behavior Inventory (GBI)

The GBI (Depue et al., 1989) is a self-report questionnaire used during the Phase I screening process to identify and distinguish between potential bipolar spectrum participants and controls. The GBI has been shown to be valid among a wide range of populations including undergraduates, psychiatric outpatients, and relatives of bipolar I probands (Depue et al, 1989; Klein, Depue, & Slater, 1985). Psychometric properties of the instrument are strong, with a reported internal consistency of a’s= 0.90–0.96, test-retest reliability of r’s= 0.71–0.74, good sensitivity (0.78), and excellent specificity (0.99) for BSDs (Depue et al., 1981, 1989). Discriminant validity also has been good (.88 in discriminating between those with and without affective disorders; Mallon, Klein, Bornstein, & Slater, 1986). The revised-GBI (Depue et al., 1989) utilized in the present study contains 73 items. Each item is designed to assess various experiences related to depressive, (hypo)manic, or biphasic symptoms, and how these experiences range on the dimensions of intensity, duration, and frequency. The respondent provides a rating on a 4-point Likert scale, ranging from 1 (not at all) to 4 (very often or almost constantly). Consistent with scoring by Depue and colleagues (1989), an item received a score of one point when rated as 3 (often) or 4 (very often or almost constantly) on the scale. Points were then summed to obtain two subscores: depression (GBI-D score) and (hypo)mania and biphasic (GBI-HB score). This scoring method provides GBI scores that represent the number of symptomatic experiences satisfying the criteria of duration, intensity, and frequency.

Based on Depue et al.’s (1989) findings, the following cutoff scores were used to identify potential bipolar spectrum and control participants: GBI-D score of ≥ 11 and GBI-HB score of ≥ 13 for potential bipolar spectrum participants; conversely, GBI-D scores < 11 and GBI-HB scores < 13 for healthy controls. A pilot study for the LIBS project validated this high-GBI and low-GBI group assignment procedure against diagnoses obtained via Schedule for Affective Disorders and Schizophrenia-Lifetime interviews (see Alloy et al., 2008).

Schedule for Affective Disorders and Schizophrenia-Lifetime (SADS-L)

The SADS-L (Endicott & Spitzer, 1978) is a semi-structured diagnostic interview with demonstrated reliability and validity to assess current and lifetime Axis I disorders. During Phase II of the screening process, diagnostic interviews were conducted using an expanded version of the SADS-L (exp-SADS-L; Alloy et al., 2008). The exp-SADS-L assesses symptoms related to mood, anxiety, eating, psychotic, and substance use disorders over the lifetime, as well as their occurrence, duration, and severity. Modifications made for use in the LIBS study include: (a) additional probes to allow for the assignment of DSM-IV-TR as well as RDC diagnoses; (b) additional items inquiring about affective symptoms developed to better capture acute characteristics of episodes as well as frequency and duration of symptoms; and (c) additional sections on eating disorders, attention-deficit hyperactivity disorder, acute stress disorder, medical history, family history, and organic rule-out conditions (Alloy et al., 2008; 2012). All interviews were conducted by extensively trained interviewers who were blind to participants’ Phase I GBI scores. In order to obtain consensus diagnoses and monitor inter-rater reliability, interviews were audiotaped. An expert psychiatric consultant was used as the third diagnostic tier for diagnostic consensus. Inter-rater reliability generated by this procedure was high: kappas for major depressive disorder diagnoses based on 80 jointly rated interviews were α > .95, while an inter-rater reliability study based on 105 jointly rated interviews yielded α > .96 for BSD diagnoses (Alloy et al., 2008).

Beck Depression Inventory (BDI)

The Beck Depression Inventory (BDI; Beck, Rush, Shaw, & Emery, 1979) is a 21–item self–report inventory that assesses the presence and severity of cognitive, motivational, affective, and somatic symptoms of depression. For the current study, a four-month version of the BDI was used, in which participants rated their symptoms for each month in the four-month interval. Thus, we averaged across the monthly scores to obtain one score representing participants’ depressive symptoms. The BDI has been shown to be valid in student samples (Bumberry, Oliver, & McClure, 1978; Hammen, 1980). Furthermore, in nonclinical populations, the internal reliability is good (α = .81 to .86) and the test–retest reliability ranges from .48 to .86, depending on the interval between re–testing and type of population (Beck, Steer, & Garbin, 1988; Groth–Marnet, 1990). In this study, the BDI had an internal consistency of α = .94. Participants completed the BDI at T1 and T2, a year later.

Halberstadt Mania Inventory (HMI)

The Halberstadt Mania Inventory (HMI; Alloy, Reilly–Harrington, Fresco, Whitehouse, & Zechmeister, 1999) is a 28–item self–report measure that assesses current (hypo)manic symptoms by covering cognitive, motivational, affective and somatic symptoms. The HMI was modeled after the BDI and has a similar format. The HMI was tested in a sample of approximately 1,282 undergraduates and had good psychometric properties, with high internal consistency (α = .82) and adequate convergent validity (r = .32, p = .001) with the mania scale of the Minnesota Multiphasic Personality Inventory (MMPI) (Hathaway & McKinley, 1951). In addition, in the LIBS project, HMI scores correlated r = .46 with hypomanic/manic symptoms rated from the expanded SADS interview. Moreover, evidence of the construct validity of the HMI has been provided by Alloy et al. (1999; 2009). In this study, the HMI had an α = .83.

Cognitive Style Questionnaire (CSQ)

The Cognitive Style Questionnaire (CSQ; Alloy et al., 2000), a modified version of the Attributional Style Questionnaire (Peterson et al., 1982; Seligman et al., 1979), measures styles for inferring causes, consequences, and characteristics about the self when faced with negative or positive life events (Abramson, Metalsky, & Alloy, 1999). The ASQ is a well-established instrument with good reliability and validity (Peterson, 1991) that assesses one's attributions for hypothetical positive and negative events on three dimensions: internality, stability, and globality. The CSQ was modified by increasing the number of events to 12 positive and 12 negative events (6 achievement and 6 interpersonal events of each valence) and by adding ratings of consequences and implications for one’s self of each event on 7-point scales. Ratings assessed the stability and globality of individuals’ causal attributions for each event as well as their inferences about the probable consequences of each event (e.g., "How likely is it that the other person no longer wanting a romantic relationship with you will lead to other negative things happening to you?") and the implications of each event for the self (e.g., "To what degree does your receiving a negative evaluation of your job performance mean to you that you are flawed in some way?"). The CSQ has exhibited excellent reliability and validity (Haeffel et al., 2008), with an internal consistency of α = .88 at T1.

Dysfunctional Attitudes Scale (DAS)

The DAS (Weissman & Beck, 1978) assesses maladaptive attitudes concerning perfectionist standards and concern about others' approval, etc. The LIBS Project used the 40-item version of the DAS Form A and added 24 items to better assess dysfunctional attitudes in the achievement and interpersonal domains. Using a 7-point Likert scale ranging from ‘Totally agree’ to ‘Totally disagree’, participants respond to descriptions of their attitudes most of the time. Factor analyses of the DAS have yielded two subscales, Approval by Others (AO – 10 items: e.g., ‘‘My value as a person depends greatly on what others think of me’’) and Performance Evaluation/Perfectionism (PE – 15 items: e.g., If I fail partly, it is as bad as being a complete failure”). The DAS has demonstrated good construct validity (Alloy et al., 2000; Francis-Raniere et al., 2006; Segal, Gemar, & Williams, 1999) and at T1 had internal consistencies for the PE and AO subscales of α = .89 and .78, respectively (Alloy et al., 2009a).

Depressive Experiences Questionnaire (DEQ)

The DEQ (Blatt et al., 1976) is a 66-item, self-report measure that measures depressive personality styles using 7-point Likert scales; responses range from strongly disagree to strongly agree. The DEQ is subdivided into three subscales: Dependency (e.g., “After an argument, I feel very lonely”), Self-Criticism (e.g., “Often, I feel as if I have disappointed others”), and Efficacy (e.g., “I am a very independent person”). The DEQ has yielded high internal and retest reliability (Blatt et al., 1976; Zuroff et al., 1983), as well as good construct validity (Zuroff et al., 2004). At T1, internal consistency for the Dependency and Self-Criticism subscales were α’s = .64 and .87, respectively (Alloy et al., 2009a). We did not use the Efficacy subscale in this study.

Sociotropy Autonomy Scale (SAS)

The SAS (Beck et al., 1983) is a 60-item questionnaire that assesses Beck’s (1987) depressive personality modes: Sociotropy and Autonomy. Sociotropy measures fears of abandonment and rejection as well as the value of interpersonal relationships. Conversely, Autonomy assesses the value one places on achievement, independence, and goal-directed behavior. Each item is rated on 5-point scales (0%, 25%, 50%, 75%, and 100%). The Sociotropy scale has high concurrent validity with numerous other measures of dependency and affiliation (Clark, Beck & Brown, 1992). The Autonomy scale is moderately correlated with another autonomy subscale, The Personality Research Form (Clark et al., 1992). The two different scales have good internal consistency (α= 0.90 and 0.93, respectively) and high retest reliability (Beck, et al., 1983; Zuroff et al., 2004). At T1, internal consistencies for Sociotropy and Autonomy were α’s = .93 and .92, respectively (Alloy et al., 2009a).

Data Analysis Approach

Initially, Pearson correlations were conducted to examine the associations between symptom levels (BDI and MHI scores) and cognitive style measures (CSQ, DAS, DEQ, and SAS; See Table 1). To test Hypotheses 1 and 2, a multivariate analysis of covariance (MANCOVA) was used to examine group differences in cognitive style measures and prospective symptomatology. This analytic technique was used because it allows for group comparisons on several related dependent measures at once, while controlling for confounds. Initial BDI and HMI symptom scores were covaried to insure that relationships between group status and cognitive styles and prospective symptoms could not be attributed to initial symptom state at T1. Further, in order to test Hypothesis 3, Hayes’ (2012) Process model, involving bootstrapping (Preacher and Hayes, 2008), was used to test whether the association between comorbid BSD and anxiety and prospective BDI and/or HMI levels could be explained by any of the cognitive styles that differentiated the BSD/Anx group from the BSD only and control groups.

Table 1.

Dependent Variable Correlations

SAS
Soc		 SAS
Aut		 DAS
PE		 DAS
AO		 CSQ
Neg		 DEQ
Dep		 DEQ
SC		 BDI
T2		 HMI
T2
SAS
Soc		 .055 .535** .663** .428** .699** .460** .236** .067
SAS
Aut		 .125 −.188* .006 −.131 .297** .139 .125
DAS
PE		 .647** .531** .402** .695** .501** −.035
DAS
AO		 .445** .557** .404** .207* .018
CSQ
Neg		 .212** .507** .192* −.107
DEQ
Dep		 .197* .306** −.005
DEQ SC .475** .009
BDI T2 .037
HMI T2
Open in a new tab
*

p<.05,

**

p<.01

Note. SAS = Sociotropy Autonomy Scale, Soc = Sociotropy, Aut = Autonomy; DAS = Dysfunctional Attitudes Scale, PE = Perfectionism subscale, AO = Approval by Others subscale; CSQ Neg = Cognitive Style Questionnaire Negative Composite; DEQ = Depressive Experiences Questionnaire, Dep = Dependency, SC = Self Criticism; BDI = Beck Depression Inventory; HMI = Halberstadt Mania Inventory, T2 = Time 2.

Results

Preliminary Analyses

First, a multivariate analysis of variance (MANOVA) was used to test whether gender or ethnicity were associated with any differences on the dependent variables of interest (DAS, DEQ, CSQ, SAS, BDI, and HMI). Wilks’s statistic revealed that there was a significant multivariate effect of ethnicity on the dependent measures, Λ= 0.38, F(4, 143)= 1.77, p <.01. Specifically, the univariate ANOVAs revealed that ethnicity had a significant effect on: SAS Sociotropy, F(4, 143)= 2.53, p<.05, CSQ, F(4, 143), p<.01, DEQ Dependency, F(4, 143)= 5.37, p<.01, DAS Perfectionism, F(4, 143)= 2.53, p< 01, DAS Approval by Others, F(4, 143)= 2.93, p<.05, and T2 BDI, F(4, 143)= 5.44), p<.01. Consequently, ethnicity was used as a covariate in the main hypothesis testing analyses.

Table 1 displays the correlations between the cognitive style and T2 symptom dependent variables. As can be seen in the table, with the exception of SAS Autonomy, most of the other cognitive style measures exhibited significant moderate intercorrelations with each other. In addition, all of the cognitive styles except SAS Autonomy were significantly correlated with depressive symptoms (BDI) at T2, whereas none of the cognitive styles correlated with hypomanic symptoms (HMI) at T2.

Group Differences in Cognitive Styles and Symptoms (Hypotheses 1 and 2)

Controlling for Time 1 BDI and HMI symptom scores, as well as ethnicity, a Multivariate Analysis of Covariance (MANCOVA) was used to compare the BSD, BSD/Anx, and Control groups on the dependent variables: DAS Perfectionism, DAS Approval by Others, DEQ Dependency, DEQ Self-Criticism, CSQ, SAS Sociotropy, SAS Autonomy all at T1, and BDI and HMI at one year follow-up (T2). Using Wilks’ statistic, there was a significant multivariate effect of group on the dependent measures, Λ= 0.52, F(2, 142)= 2.32, p <.01. Further, the univariate ANCOVAs revealed significant differences between the three groups on: DAS Perfectionism, DEQ Dependency, DEQ Self-Criticism, SAS Autonomy, SAS Sociotropy, CSQ, and T2 BDI. The groups were not significantly different on DAS Approval by Others and T2 HMI (See Table 2). Although the multivariate test and follow-up univariate analyses inform us of the significance of overall group differences, which particular groups differed significantly from each other on each dependent variable was unclear.

Table 2.

Descriptive Statistics of Dependent Measures

BSD/Anx BSD Control F
Dependent
Measure		 Mean SD Mean SD Mean SD
SAS Soc 97.04 19.76 85.19 19.39 82.66 13.52 8.49***
SAS Aut 106.98 11.90 103.32 15.33 96.92 14.46 5.98**
DAS PE 3.33 1.33 2.53 0.91 2.19 0.76 12.44***
DAS AO 3.95 1.19 3.49 1.00 3.48 0.79 2.99
CSQ 4.23 1.23 3.66 0.82 3.11 1.00 11.71***
DEQ Dep −0.52 0.96 −1.19 0.94 −0.89 0.78 6.71**
DEQ SC 0.59 1.01 −0.41 1.09 −1.03 0.87 29.00***
BDI T2 5.96 6.11 2.82 3.68 0.83 1.19 14.56***
HMI T2 11.37 5.90 10.04 4.62 10.55 4.98 0.81
Open in a new tab
*

p<.05,

**

p<.01,

***

p<.001

Note. F-statistic denotes significant multivariate differences between all three groups. SAS = Sociotropy Autonomy Scale, Soc = Sociotropy, Aut = Autonomy; DAS = Dysfunctional Attitudes Scale, PE = Perfectionism subscale, AO = Approval by Others subscale; CSQ Neg = Cognitive Style Questionnaire Negative Composite; DEQ = Depressive Experiences Questionnaire, Dep = Dependency, SC = Self Criticism; BDI = Beck Depression Inventory; HMI = Halberstadt Mania Inventory, T2 = Time 2.

In order to determine the specific group differences, a series of planned comparisons was conducted. A Bonferroni correction was used to adjust the significance level. Planned contrasts revealed that the BSD/Anx group scored higher than the BSD group on SAS Sociotropy, DEQ Dependency, DEQ Self-Criticism, DAS Perfectionism, and T2 BDI (See Table 3). Further, the BSD/Anx group scored higher than the Control group on SAS Sociotropy, SAS Autonomy, CSQ, DAS Perfectionism, DAS Approval by Others, and T2 BDI. Planned contrasts revealed no group differences on DEQ Dependency and T2 HMI (See Table 4). These observed group differences are meaningful because the interval boundaries do not cross zero.

Table 3.

Planned Comparisons between BSD and BSD/Anx Groups

Groups t Sig. (2-tailed) 95% CI
SAS Soc BD −2.80 .006 −18.03, −3.07
BD/Anx
SAS Aut BD −1.96 .053 −10.68, 0.07
BD/Anx
DAS PE BD −3.77 .000 −1.18, −0.37
BD/Anx
CSQ Neg BD −1.66 .100 −0.86, 0.08
BD/Anx
DEQ_Dep BD −3.84 .000 −1.02, −0.33
BD/Anx
DEQ_SC BD −4.97 .000 −1.40, −0.60
BD/Anx
BDI T2 BD −3.47 .001 −5.65, −1.53
BD/Anx
Open in a new tab

Note. SAS = Sociotropy Autonomy Scale, Soc = Sociotropy, Aut = Autonomy; DAS = Dysfunctional Attitudes Scale, PE = Perfectionism subscale; CSQ Neg = Cognitive Style Questionnaire Negative Composite; DEQ = Depressive Experiences Questionnaire, Dep = Dependency, SC = Self Criticism; BDI = Beck Depression Inventory; T2 = Time 2.

Table 4.

Planned Comparisons between Control and BSD/Anx Groups

Groups t Sig. (2-tailed) 95% CI
SAS Soc Control −2.83 .006 −16.10, −2.84
BD/Anx
SAS Aut Control −4.13 .000 −14.41, −5.07
BD/Anx
DAS PE Control −3.93 .000 −1.22, −0.40
BD/Anx
CSQ Neg Control −2.17 .032 −1.04, −0.05
BD/Anx
DEQ SC Control −5.73 .000 −1.58, −0.77
BD/Anx
DEQ Dep Control −1.94 .055 −0.61, 0.01
BD/Anx
BDI T2 Control −5.61 .000 −7.26, −3.45
BD/Anx
Open in a new tab

Note. SAS = Sociotropy Autonomy Scale, Soc = Sociotropy, Aut = Autonomy; DAS = Dysfunctional Attitudes Scale, PE = Perfectionism subscale; CSQ Neg = Cognitive Style Questionnaire Negative Composite; DEQ = Depressive Experiences Questionnaire, Dep = Dependency, SC = Self Criticism; BDI = Beck Depression Inventory; T2 = Time 2.

Cognitive Styles as Mediators of the Association Between Bipolar-Anxiety Comorbidity and Prospective Symptoms

Hayes' (2012) Process was used to test for mediational effects of SAS Sociotropy, DEQ Dependency, DEQ Self-Criticism, and DAS Perfectionism on the positive relationship between BSD/Anx group and higher prospective depressive symptoms. Group (BSD or BSD/Anx) was entered as the independent variable, with T2 BDI entered as the dependent variable. Ethnicity and T1 BDI were entered as covariates. SAS Sociotropy, DEQ Dependency, DEQ Self-Criticism, and DAS Perfectionism were entered as possible mediators. The 95% confidence intervals for the indirect effect of group on T2 depressive symptoms through DAS Perfectionism did not include zero, indicating the presence of a significant mediating relationship. The 95% confidence intervals for the indirect effect of group on T2 depressive symptoms through SAS Sociotropy, DEQ Dependency, and DEQ Self-Criticism did include zero, indicating no significant mediating relationships for these other cognitive styles (See Table 5).

Table 5.

Bootstrapping test of mediation

T2 BDI
Predictor Direct Effect of X on Y Indirect Effect of X on Y BootLLCI BootUCLI
Group 1.353* SAS Soc 0.009 −0.45 0.564
DAS PE 1.982* 0.490 3.968
DEQ Dep 0.180 −0.01 0.588
DEQ SC 0.348 −0.205 1.012
Open in a new tab
*

p<.05.

Note. SAS = Sociotropy Autonomy Scale, Soc = Sociotropy, DAS = Dysfunctional Attitudes Scale, PE = Perfectionism subscale; DEQ = Depressive Experiences Questionnaire, Dep = Dependency, SC = Self Criticism.

Discussion

Researchers have attempted to understand why comorbidity between mood and anxiety disorders results in more deleterious outcomes than mood disorders alone. Even after taking into account initial levels of depressive and hypomanic/manic symptoms, we found that those with BSD/anxiety comorbidity experienced more severe depressive symptoms, but not more severe hypomanic/manic symptoms, over a one-year follow-up. The results of the present analyses also indicated quite clearly that individuals with BSD/anxiety comorbidity interpreted their environments with cognitive styles that are more negative than those with BSDs only or healthy controls. These cognitive styles were characterized by more extreme sociotropy, dependency, self-criticism, and perfectionism. However, only perfectionistic attitudes mediated the association between BSD and anxiety comorbidity and increased depressive symptoms at one-year follow-up.

In addition to the evidence that BSD/anxiety comorbidity results in exacerbated depressive symptoms over time, we found that a certain cognitive style contributes to this worse depressive course, specifically, a depressive style grounded in high perfectionism. Individuals with both a BSD and anxiety diagnosis engaged in more perfectionistic thinking than individuals with a BSD only. This perfectionistic thinking style engendered more severe depressive symptomatology. This study is relatively unique in that, to date, little research has been conducted on the possible mechanisms that underlie the more negative course of disorder associated with bipolar-anxiety comorbidity.

Perfectionism may be especially likely to mediate the comorbidity-depression relationship because of its integral association with both anxiety and depression. Perfectionism has been implicated in the development and maintenance of various anxiety disorders (Antony et al., 1998). Specifically, cognitive theories of obsessive compulsive disorder (OCD) assert that perfectionistic thinking underlies one’s obsessions. These theories have been supported by Frost and Steketee (1997) and later applied to panic disorder (Antony and Barlow, 1996). Individuals with panic disorder endorsed higher scores on measures that assessed perfectionism and concern about one’s mistakes than non-anxious controls (Antony et al., 1992; Antony and Barlow, 1996; Antony et al., 1998). Similarly, perfectionistic thinking has been shown to be present in depressed (Kawamura et al., 2001) and BSD individuals (Scott et al., 2000). Moreover, perfectionism consistently has been associated with severe depressive symptoms (Hewitt and Flett, 1990; Hewitt and Flett, 1991; Cox, and Enns, 2003). Given previous research on perfectionism, the finding that it mediated the BSD/anxiety comorbidity- worse depressive symptom relationship is not surprising.

Interestingly, BSD/anxiety comorbidity did not seem to lead to worse (hypo)manic symptoms in our study. Although these results are in line with most of the literature (Simon et al., 2004; Otto et al., 2006; Simon et al., 2007), Akiskal et al. (2006) found that panic-agoraphobia comorbidity was associated with more frequent hypomanic episodes. To date, there is not a clear understanding of why manic/hypomanic symptoms sometimes are and sometimes are not aggravated by BSD/anxiety comorbidity. Future research should seek to understand the conditions under which manic/hypomanic outcomes are affected.

This study is consistent with the research of Simon and colleagues (2007) that comorbid anxiety disorders among individuals with bipolar disorder stimulate worse depressive symptoms. Moreover, our results are consistent with previous literature that demonstrates that dysfunctional cognitive styles are observed in BSD individuals (Alloy et al., 2009a). Earlier work by Burns and Spangler (2001) and Sanz and Avia (1994) demonstrated that both depressed and anxious individuals endorse distinct negative cognitive styles. Expanding on these findings, the present study illustrates the negative cognitive styles that are even more extreme in individuals who exhibit both BSDs and anxiety disorders. Additionally, the results of the present study clarify work by Simon et al. (2007), Goes et al. (2012), and Coryell et al. (2012) by providing a more detailed explanation for the processes by which comorbidity produces depressive exacerbation.

Among the strengths of this study were the relatively large and equal sample sizes of the three groups of participants, the inclusion of demographically matched controls, the examination of a number of cognitive styles previously associated with depression and anxiety, and the prediction of prospective mood symptoms. These qualities afforded an ability to measure the impact of anxiety comorbidity on depressive symptoms as well as some of the cognitive mechanisms that may help to explain this impact. Despite these strengths, the study had several limitations. First, although this study was longitudinal, there were only two waves of data. It would be more advantageous to use three waves of data as it would allow one to investigate the full causal effect of one variable on another (Cole and Maxwell, 2003). Second, in this study, a comorbid anxiety diagnosis consisted of any DSM-IV-TR or RDC anxiety diagnosis. If individuals were separated by their specific anxiety diagnoses, this could afford a more complete understanding of the effect of types of anxiety on prospective depressive symptoms. However, a much larger sample size would be needed to do this. A similar criticism is relevant to the bipolar groups. The groups were comprised of participants who met criteria for any BSD except Bipolar I. In order to examine unique effects of comorbidity on BSDs, future research should use groups that are strictly separated by diagnosis (e.g., cyclothymia, BD II, BD I). Another limitation is that we only examined the effects of anxiety comorbidity on the course of BSDs for self-reported depressive and hypomanic/manic symptoms. Future research should examine whether maladaptive cognitive styles also help to account for the effects of anxiety comorbidity on BSD individuals’ suicidality, functional impairment, and diagnosed mood episode onset.

Perhaps the most important lesson to be learned from this study is that a depressogenic cognitive style involving perfectionism leads to more severe depressive symptomatology. This distinctive cognitive style may be especially pronounced in individuals with bipolar and anxiety comorbidity and be one of the underlying mechanisms that contributes to individuals with BSD/Anx comorbidity experiencing more severe depressive symptoms. Future research should examine what, if any, anxiety-specific attitudes are adaptive in preventing an aggravation of hypo(manic) symptoms or episodes in individuals with both a bipolar and an anxiety disorder.

Acknowledgments

This research was supported by National Institute of Mental Health grants MH 52617, MH 77908, and MH 102310 to Lauren B. Alloy and MH 52662 to Lyn Y. Abramson.

Contributor Information

Jared K. O’Garro-Moore, Temple University

Ashleigh Molz Adams, Temple University.

Lyn Y. Abramson, University of Wisconsin-Madison

Lauren B. Alloy, Temple University

References

  1. Abramson LY, Metalsky GI, Alloy LB. Hopelessness depression: A theory-based subtype of depression. Psychological review. 1989;96(2):358–372. [Google Scholar]
  2. Akiskal HS, Akiskal KK, Perugi G, Toni C, Ruffolo G, Tusini G. Bipolar II and anxious reactive “comorbidity”: Toward better phenotypic characterization suitable for genotyping. Journal of Affective Disorders. 2006;96(3):239–247. doi: 10.1016/j.jad.2006.08.010. [DOI] [PubMed] [Google Scholar]
  3. Alford BA, Gerrity DM. The specificity of sociotropy-autonomy personality dimensions to depression vs. anxiety. Journal of Clinical Psychology. 1995;51(2):190–195. doi: 10.1002/1097-4679(199503)51:2<190::aid-jclp2270510207>3.0.co;2-s. [DOI] [PubMed] [Google Scholar]
  4. Alloy LB, Abramson LY, Whitehouse WG, Hogan ME, Tashman NA, L Steinberg D, Donovan P. Depressogenic cognitive styles: Predictive validity, information processing and personality characteristics, and developmental origins. Behaviour Research and Therapy. 1999;37(6):503–531. doi: 10.1016/s0005-7967(98)00157-0. [DOI] [PubMed] [Google Scholar]
  5. Alloy LB, Abramson LY, Smith JM, Gibb BE, Neeren AM. Role of parenting and maltreatment histories in unipolar and bipolar mood disorders: Mediation by cognitive vulnerability to depression. Clinical Child and Family Psychology Review. 2006;9(1):23–64. doi: 10.1007/s10567-006-0002-4. [DOI] [PubMed] [Google Scholar]
  6. Alloy LB, Abramson LY, Walshaw PD, Cogswell A, Grandin LD, Hughes ME, Hogan ME. Behavioral Approach System and Behavioral Inhibition System sensitivities and bipolar spectrum disorders: prospective prediction of bipolar mood episodes. Bipolar Disorders. 2008;10(2):310–322. doi: 10.1111/j.1399-5618.2007.00547.x. [DOI] [PubMed] [Google Scholar]
  7. Alloy LB, Abramson LY, Walshaw PD, Gernstein RK, Keyser JD, Whitehouse WG, Urosevic S, Nusslock R, Hogan ME, Harmon-Jones E. Behavioral Approach System (BAS)- relevant cognitive styles and bipolar spectrum disorders: Concurrent and prospective associations. Journal of Abnormal Psychology. 2009a;118(3):459–471. doi: 10.1037/a0016604. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Alloy LB, Bender RE, Wagner CA, Whitehouse WG, Abramson LY, Hogan ME, Sylvia LG, Harmon-Jones E. Bipolar spectrum – substance use co-occurrence: Behavioral Approach System (BAS) sensitivity and impulsiveness as shared personality vulnerabilities. Journal of Personality and Social Psychology. 2009b;97:549–565. doi: 10.1037/a0016061. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Alloy LB, Reilly-Harrington N, Fresco DM, Whitehouse WG, Zechmeister JS. Cognitive styles and life events in subsyndromal unipolar and bipolar mood disorders: Stability and prospective prediction of depressive and hypomanic mood swings. Journal of Cognitive Psychotherapy: An International Quarterly. 1999;13:21–40. [Google Scholar]
  10. Alloy LB, Urošević S, Abramson LY, Jager-Hyman S, Nusslock R, Whitehouse WG, Hogan M. Progression along the bipolar spectrum: A longitudinal study of predictors of conversion from bipolar spectrum conditions to bipolar I and II disorders. Journal of Abnormal Psychology. 2012;121(1):16–27. doi: 10.1037/a0023973. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. American Psychiatric Association. Diagnostic and statistical manual. 4th. Washington, D.C.: Author; 2000. Text Revised. [Google Scholar]
  12. Angst F, Stassen HH, Clayton PJ, Angst J. Mortality of patients with mood disorders: follow-up over 34–38 years. Journal of Affective Disorders. 2002;68(2):167–181. doi: 10.1016/s0165-0327(01)00377-9. [DOI] [PubMed] [Google Scholar]
  13. Angst J, Angst F, Gerber-Werder R, Gamma A. Suicide in 406 mood disorder patients with and without long-term medication: a 40 to 44 years’ follow-up. Archives of Suicide Research. 2005;9:279–300. doi: 10.1080/13811110590929488. [DOI] [PubMed] [Google Scholar]
  14. Antony MM, Brown TA, Barlow DH. Current perspectives on panic and panic disorder. Current Directions in Psychological Science. 1992;1:79–82. [Google Scholar]
  15. Antony MM, Barlow DH. Emotion theory as a framework for explaining panic attacks and panic disorder. In: Rapee RM, editor. Current controversies in the anxiety disorders. New York, NY: Guilford Publications; 1996. pp. 55–76. [Google Scholar]
  16. Antony MM, Bieling PJ, Cox BJ, Enns MW, Swinson RP. Psychometric properties of the 42-item and 21-item versions of the Depression Anxiety Stress Scales in clinical groups and a community sample. Psychological Assessment. 1998;10(2):176–181. [Google Scholar]
  17. Beck AT, Rush AJ, Shaw BF, Emery G. Cognitive therapy of depression. New York: Guilford Press; 1979. [Google Scholar]
  18. Beck AT, Epstein N, Harrison R. Cognitions, attitudes and personality dimensions in depression. British Journal of Cognitive Psychotherapy. 1983;1(1):1–16. [Google Scholar]
  19. Beck AT, Brown G, Steer RA, Eidelson JI, Riskind JH. Differentiating anxiety and depression: a test of the cognitive content-specificity hypothesis. Journal of Abnormal Psychology. 1987;96(3):179–183. doi: 10.1037//0021-843x.96.3.179. [DOI] [PubMed] [Google Scholar]
  20. Beck AT, Steer RS, Garbin M. The Beck Depression Inventory: Twenty-five years of evaluation. Clinical Psychology Review. 1988;8:77–100. [Google Scholar]
  21. Blatt SJ. Levels of object representation in anaclitic and introjective depression. Psychoanalytic Study of the Child. 1974;29:107–157. [PubMed] [Google Scholar]
  22. Blatt SJ, D'Afflitti JP, Quinlan DM. Experiences of depression in normal young adults. Journal of Abnormal Psychology. 1976;85(4):383–389. doi: 10.1037//0021-843x.85.4.383. [DOI] [PubMed] [Google Scholar]
  23. Bögels SM, Zigterman D. Dysfunctional cognitions in children with social phobia, separation anxiety disorder, and generalized anxiety disorder. Journal of Abnormal Child Psychology. 2000;28(2):205–211. doi: 10.1023/a:1005179032470. [DOI] [PubMed] [Google Scholar]
  24. Brenning K, Soenens B, Braet C, Beyers W. Longitudinal dynamics of depressogenic personality and attachment dimensions in adolescence: An examination of associations with changes in depressive symptoms. Journal of Youth and Adolescence. 2013;42(8):1128–1144. doi: 10.1007/s10964-012-9879-z. [DOI] [PubMed] [Google Scholar]
  25. Brown EJ, Juster HR, Heimberg RG, Winning CD. Stressful life events and personality styles: Relation to impairment and treatment outcome in patients with social phobia. Journal of Anxiety Disorders. 1998;12(3):233–251. doi: 10.1016/s0887-6185(98)00012-7. [DOI] [PubMed] [Google Scholar]
  26. Bumberry W, Oliver JM, McClure JN. Validation of the Beck Depression Inventory in a university population using psychiatric estimate as the criterion. Journal of Consulting and Clinical Psychology. 1978;46(1):150–155. [Google Scholar]
  27. Burns DD, Spangler DL. Do changes in dysfunctional attitudes mediate changes in depression and anxiety in cognitive behavioral therapy? Behavior Therapy. 2001;32(2):337–369. [Google Scholar]
  28. Clark DA, Beck AT. Personality factors in dysphoria: A psychometric refinement of Beck's Sociotropy-Autonomy Scale. Journal of Psychopathology and Behavioral Assessment. 1991;13(4):369–388. [Google Scholar]
  29. Clark DA, Beck AT, Brown GK. Sociotropy, autonomy, and life event perceptions in dysphoric and nondysphoric individuals. Cognitive Therapy and Research. 1992;16(6):635–652. [Google Scholar]
  30. Cohen JR, Young JF, Hankin BL, Yao S, Zhu XZ, Abela JR. Personality predispositions in Chinese adolescents: The relation between self-criticism, dependency, and prospective internalizing symptoms. Journal of Social and Clinical Psychology. 2013;32(6):596–618. doi: 10.1521/jscp.2013.32.6.596. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Cole DA, Maxwell SE. Testing mediational models with longitudinal data: Questions and tips in the use of structural equation modeling. Journal of Abnormal Psychology. 2003;112(4):558–577. doi: 10.1037/0021-843X.112.4.558. [DOI] [PubMed] [Google Scholar]
  32. Coryell W, Fiedorowicz JG, Solomon D, Leon AC, Rice JP, Keller MB. Effects of anxiety on the long-term course of depressive disorders. The British Journal of Psychiatry. 2012;200(3):210–215. doi: 10.1192/bjp.bp.110.081992. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Cox BJ, Enns MW. Relative stability of dimensions of perfectionism in depression. Canadian Journal of Behavioural Science/Revue Canadienne des Sciences du Comportement. 2003;35(2):124–132. [Google Scholar]
  34. Depue RA, Slater JF, Wolfstetter-Kausch H, Klein D, Goplerud E, Farr D. A behavioral paradigm for identifying persons at risk for bipolar depressive disorder: a conceptual framework and five validation studies. Journal of Abnormal Psychology. 1981;90(5):381–437. doi: 10.1037//0021-843x.90.5.381. [DOI] [PubMed] [Google Scholar]
  35. Depue RA, Iacono WG. Neurobehavioral aspects of affective disorders. Annual Review of Psychology. 1989;40(1):457–492. doi: 10.1146/annurev.ps.40.020189.002325. [DOI] [PubMed] [Google Scholar]
  36. Endicott J, Spitzer RL. A diagnostic interview: The schedule for affective disorders and schizophrenia. Archives of General Psychiatry. 1978;35(7):837–844. doi: 10.1001/archpsyc.1978.01770310043002. [DOI] [PubMed] [Google Scholar]
  37. Flett GL, Hewitt PL, Dyck DG. Self-oriented perfectionism, neuroticism and anxiety. Personality and Individual Differences. 1989;10(7):731–735. [Google Scholar]
  38. Francis-Raniere EL, Alloy LB, Abramson LY. Depressive personality styles and bipolar spectrum disorders: Prospective tests of the event congruency hypothesis. Bipolar Disorders. 2006;8(4):382–399. doi: 10.1111/j.1399-5618.2006.00337.x. [DOI] [PubMed] [Google Scholar]
  39. Fresco DM, Sampson WS, Craighead LW, Koons AN. The relationship of sociotropy and autonomy to symptoms of depression and anxiety. Journal of Cognitive Psychotherapy. 2001;15(1):17–31. [Google Scholar]
  40. Frost RO, Steketee G. Perfectionism in obsessive–compulsive disorder patients. Behaviour Research and Therapy. 1997;35(4):291–296. doi: 10.1016/s0005-7967(96)00108-8. [DOI] [PubMed] [Google Scholar]
  41. Goes FS, McCusker MG, Bienvenu OJ, Mackinnon DF, Mondimore FM, Schweizer B, DePaulo JR, Potash JB. Comorbid anxiety disorders in bipolar disorder and major depression: Familial aggregation and clinical characteristics of comorbid panic disorder, social phobia, specific phobia and obsessive-compulsive disorder. Psychological Medicine. 2012;42:1449–1459. doi: 10.1017/S0033291711002637. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Goodwin FK, Jamison KR. Manic-depressive illness: Bipolar disorder and recurrent depression. New York, NY: Oxford University Press; 2007. [Google Scholar]
  43. Groth-Marnat G. Handbook of Psychological Assessment. New York, NY: John Wiley & Sons; 2009. [Google Scholar]
  44. Hammen CL. Depression in college students: Beyond the Beck Depression Inventory. Journal of Consulting and Clinical Psychology. 1980;48(1):126–128. doi: 10.1037//0022-006x.48.1.126. [DOI] [PubMed] [Google Scholar]
  45. Harrison EC, Barraclough B. Suicide as an outcome for mental disorders: A meta-analysis. The British Journal of Psychiatry. 1997;170(3):205–228. doi: 10.1192/bjp.170.3.205. [DOI] [PubMed] [Google Scholar]
  46. Hathaway SR, McKinley JC. Minnesota Multiphasic Personality Inventory; manual (Revised) San Antonio, TX: Psychological Corporation; 1951. p. 31. [Google Scholar]
  47. Hayes AF. PROCESS: A versatile computational tool for observed variable mediation, moderation, and conditional process modeling. 2012 [White paper]. Retrieved from http://www.afhayes.com/public/process2012.pdf. [Google Scholar]
  48. Heimberg RG, Klosko JS, Dodge CS, Shadick R, Becker RE, Barlow DH. Anxiety disorders, depression, and attributional style: A further test of the specificity of depressive attributions. Cognitive Therapy and Research. 1989;13(1):21–36. [Google Scholar]
  49. Henry C, Van den Bulke D, Bellivier F, Etain B, Rouillon F, Leboyer M. Anxiety disorders in 318 bipolar patients: Prevalence and impact on illness severity and response to mood stabilizer. Journal of Clinical Psychiatry. 2003;64(3):331–335. [PubMed] [Google Scholar]
  50. Hewitt PL, Flett GL. Perfectionism and depression: A multidimensional analysis. Journal of Social Behavior & Personality. 1990;5(5):423–438. [Google Scholar]
  51. Hewitt PL, Flett GL. Dimensions of perfectionism in unipolar depression. Journal of Abnormal Psychology. 1991;100(1):98–101. doi: 10.1037//0021-843x.100.1.98. [DOI] [PubMed] [Google Scholar]
  52. Hirschfeld R, Klerman GL, Clayton PJ, Keller MB. Personality and depression: Empirical findings. Archives of General Psychiatry. 1983;40(9):993–998. doi: 10.1001/archpsyc.1983.01790080075010. [DOI] [PubMed] [Google Scholar]
  53. Kessler RC, Berglund P, Demler O, Jin R, Koretz D, Merikangas KR, Wang PS. The epidemiology of major depressive disorder: Results from the National Comorbidity Survey Replication (NCS-R) JAMA. 2003;289(23):3095–3105. doi: 10.1001/jama.289.23.3095. [DOI] [PubMed] [Google Scholar]
  54. Klein DN, Depue RA, Slater JF. Cyclothymia in the adolescent offspring of parents with bipolar affective disorder. Journal of Abnormal Psychology. 1985;94(2):115–127. doi: 10.1037//0021-843x.94.2.115. [DOI] [PubMed] [Google Scholar]
  55. Mallon JC, Klein DN, Bornstein RF, Slater JF. Discriminant validity of the General Behavior Inventory: An outpatient study. Journal of Personality Assessment. 1986;50(4):568–577. doi: 10.1207/s15327752jpa5004_4. [DOI] [PubMed] [Google Scholar]
  56. Masih S, Spence SH, Oei TP. Sociotropic and autonomous personality and stressful life events as predictors of depressive symptoms in the postpartum period. Cognitive Therapy and Research. 2007;31(4):483–502. [Google Scholar]
  57. McElroy SL, Altshuler LL, Suppes T, Keck PE, Frye MA, Denicoff KD, Post RM. Axis I psychiatric comorbidity and its relationship to historical illness variables in 288 patients with bipolar disorder. American Journal of Psychiatry. 2001;158(3):420–426. doi: 10.1176/appi.ajp.158.3.420. [DOI] [PubMed] [Google Scholar]
  58. Merikangas KR, Akiskal HS, Angst J, Greenberg PE, Hirschfeld RM, Petukhova M, Kessler RC. Lifetime and 12-month prevalence of bipolar spectrum disorder in the National Comorbidity Survey replication. Archives of General Psychiatry. 2007;64:543–552. doi: 10.1001/archpsyc.64.5.543. [DOI] [PMC free article] [PubMed] [Google Scholar]
  59. Miklowitz DJ, Johnson SL. The psychopathology and treatment of bipolar disorder. Annual Review of Clinical Psychology. 2006;2:199–235. doi: 10.1146/annurev.clinpsy.2.022305.095332. [DOI] [PMC free article] [PubMed] [Google Scholar]
  60. Otto MW, Simon NM, Wisniewski SR, Miklowitz DJ, Kogan JN, Reilly-Harrington NA, Pollack MH. Prospective 12-month course of bipolar disorder in out-patients with and without comorbid anxiety disorders. The British Journal of Psychiatry. 2006;189(1):20–25. doi: 10.1192/bjp.bp.104.007773. [DOI] [PubMed] [Google Scholar]
  61. Parker G, McCraw S, Fletcher K. Cyclothymia. Depression and Anxiety. 2012;29(6):487–494. doi: 10.1002/da.21950. [DOI] [PubMed] [Google Scholar]
  62. Perroud N, Baud P, Preisig M, Etain B, Bellivier F, Favre S, Malafosse A. Social phobia is associated with suicide attempt history in bipolar inpatients. Bipolar Disorders. 2007;9(7):713–721. doi: 10.1111/j.1399-5618.2007.00471.x. [DOI] [PubMed] [Google Scholar]
  63. Peterson C, Semmel A, von Baeyer C, Abramson LY, Metalsky GI, Seligman ME. The attributional style questionnaire. Cognitive Therapy and Research. 1982;6(3):287–299. [Google Scholar]
  64. Peterson C. The meaning and measurement of explanatory style. Psychological Inquiry. 1991;2(1):1–10. [Google Scholar]
  65. Pini S, Dell’Osso L, Amador XF, Mastrocinque C, Saettoni M, Cassano GB. Awareness of illness in patients with bipolar I disorder with or without comorbid anxiety disorders. Australian and New Zealand Journal of Psychiatry. 2003;37(3):355–361. doi: 10.1046/j.1440-1614.2003.01188.x. [DOI] [PubMed] [Google Scholar]
  66. Provencher MD, Guimond AJ, Hawke LD. Comorbid anxiety in bipolar spectrum disorders: A neglected research and treatment issue? Journal of Affective Disorders. 2012;137(1):161–164. doi: 10.1016/j.jad.2011.12.001. [DOI] [PubMed] [Google Scholar]
  67. Reilly-Harrington NA, Alloy LB, Fresco DM, Whitehouse WG. Cognitive styles and life events interact to predict bipolar and unipolar symptomatology. Journal of Abnormal Psychology. 1999;108(4):567–578. doi: 10.1037//0021-843x.108.4.567. [DOI] [PubMed] [Google Scholar]
  68. Sala R, Goldstein BI, Morcillo C, Liu SM, Castellanos M, Blanco C. Course of comorbid anxiety disorders among adults with bipolar disorder in the US population. Journal of Psychiatric Research. 2012;46(7):865–872. doi: 10.1016/j.jpsychires.2012.03.024. [DOI] [PMC free article] [PubMed] [Google Scholar]
  69. Sanz J, Avia MD. Cognitive specificity in social anxiety and depression: Self-statements, self-focused attention, and dysfunctional attitudes. Journal of Social and Clinical Psychology. 1994;13(2):105–137. [Google Scholar]
  70. Saunders EF, Fitzgerald KD, Zhang P, McInnis MG. Clinical features of bipolar disorder comorbid with anxiety disorders differ between men and women. Depression and Anxiety. 2012;29(8):739–746. doi: 10.1002/da.21932. [DOI] [PMC free article] [PubMed] [Google Scholar]
  71. Scott J, Stanton B, Garland A, Ferrier IN. Cognitive vulnerability in patients with bipolar disorder. Psychological Medicine. 2000;30(2):467–472. doi: 10.1017/s0033291799008879. [DOI] [PubMed] [Google Scholar]
  72. Segal ZV, Gemar M, Williams S. Differential cognitive response to a mood challenge following successful cognitive therapy or pharmacotherapy for unipolar depression. Journal of Abnormal Psychology. 1999;108(1):3–10. doi: 10.1037//0021-843x.108.1.3. [DOI] [PubMed] [Google Scholar]
  73. Shahar G, Soffer N, Gilboa-Shechtman E. Sociotropy, autonomy, and self-criticism are three distinguishable dimensions of cognitive-personality vulnerability. Journal of Cognitive Psychotherapy. 2008;22(3):219–227. [Google Scholar]
  74. Simon NM, Otto MW, Wisniewski SR, Fossey M, Sagduyu K, Frank E, Pollack MH. Anxiety disorder comorbidity in bipolar disorder patients: Data from the first 500 participants in the Systematic Treatment Enhancement Program for Bipolar Disorder (STEP-BD) American Journal of Psychiatry. 2004;161(12):2222–2229. doi: 10.1176/appi.ajp.161.12.2222. [DOI] [PubMed] [Google Scholar]
  75. Simon NM, Zalta AK, Otto MW, Ostacher MJ, Fischmann D, Chow CW, Pollack MH. The association of comorbid anxiety disorders with suicide attempts and suicidal ideation in outpatients with bipolar disorder. Journal of Psychiatric Research. 2007;41(3):255–264. doi: 10.1016/j.jpsychires.2006.08.004. [DOI] [PubMed] [Google Scholar]
  76. Spitzer RL, Endicott J, Robins E. Research diagnostic criteria: Rationale and reliability. Archives of General Psychiatry. 1978;35(6):773–782. doi: 10.1001/archpsyc.1978.01770300115013. [DOI] [PubMed] [Google Scholar]
  77. Sutton JM, Mineka S, Zinbarg RE, Craske MG, Griffith JW, Rose RD, Mor N. The relationships of personality and cognitive styles with self-reported symptoms of depression and anxiety. Cognitive Therapy and Research. 2011;35(4):381–393. doi: 10.1007/s10608-010-9336-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  78. Zimmerman M, Chelminski I, McDermut W. Major depressive disorder and axis I diagnostic comorbidity. Journal of Clinical Psychiatry. 2002;63:187–193. doi: 10.4088/jcp.v63n0303. [DOI] [PubMed] [Google Scholar]
  79. Zuroff DC, Moskowitz DS, Wielgus MS, Powers TA, Franko DL. Construct validation of the dependency and self-criticism scales of the Depressive Experiences Questionnaire. Journal of Research in Personality. 1983;17(2):226–241. [Google Scholar]
  80. Zuroff DC, Mongrain M, Santor DA. Conceptualizing and measuring personality vulnerability to depression: Comment on Coyne and Whiffen (1995) Psychological Bulletin. 2004;130(3):489–511. doi: 10.1037/0033-2909.130.3.489. [DOI] [PubMed] [Google Scholar]

RESOURCES