A Pathoplastic Vulnerability Model: an Association Between Traumatic Stressful Life Events & OCD Kiara R

Florida State University Libraries

Electronic Theses, Treatises and Dissertations The Graduate School

2005 A Pathoplastic Vulnerability Model: An Association Between Traumatic Stressful Life Events & OCD Kiara R. Cromer

Follow this and additional works at the FSU Digital Library. For more information, please contact [email protected] THE FLORIDA STATE UNIVERSITY

COLLEGE OF ARTS AND SCIENCES

A PATHOPLASTIC VULNERABILITY MODEL:

AN ASSOCIATION BETWEEN TRAUMATIC STRESSFUL LIFE EVENTS & OCD

KIARA R. CROMER

A Thesis submitted to the Department of Psychology in partial fulfillment of the requirements for the degree of Master of Science

Degree Awarded: Fall Semester, 2005

The members of the Committee approve the thesis of Kiara Cromer defended on August 1, 2005.

______Brad Schmidt Professor Directing Thesis

______Thomas Joiner Committee Member

______Chris Schatschneider Committee Member

The Office of Graduate Studies has verified and approved the above named committee members.

TABLE OF CONTENTS

List of Tables ……………………………………………………………………………… iv List of Figures …………………………………………………………………………….. v List of Abbreviations ……………………………………………………………………... vi Abstract …………………………………………………………………………………… vii

INTRODUCTION ………………………………………………………………………… 1

Stressful Life Events and Anxiety Disorders ……………………………………… 1

Stressful Life Events and OCD ……………………………………………………. 3

Stressful Life Events and the onset of OCD ………………………………………. 4

Stressful Life Events and the course of OCD ……………………………………... 6

Stressful Life Events that Occur in Childhood vs. Adulthood ……………………..6

Models of Vulnerability Factors in Psychopathology …………………………….. 8

Present Study Aims and Hypotheses ……………………………………………… 9

METHODS ………………………………………………………………………………... 11

Participants ………………………………………………………………………… 11

Measures …………………………………………………………………………... 11

Data Analytic Plan ………………………………………………………………… 14

RESULTS …………………………………………………………………………………. 16

DISCUSSION …………………………………………………………………………….. 20

FOOTNOTES …………………………………………………………………………….. 25

APPENDICES ……………………………………………………………………………. 36

REFERENCES ……………………………………………………………………………. 38

BIOGRAPHICAL SKETCH ……………………………………………………………… 46

iii

LIST OF TABLES

1. Intercorrelations Between OCD Variables and SLE …………………………………... 26

2. Summary of Hierarchical Regression Analyses for SLE Predicting OCD …………….. 27 Symptom Severity

3. Comparison of the Four Symptom Dimensions ……………………………………….. 28

4. Summary of Hierarchical Regression Analyses for SLE Predicting OCD …………….. 29 Symptom Dimensions

5. Summary of Hierarchical Regression Analyses for SLE Predicting Comorbidity …….. 30

6. Comparison of the three SLE Measures Based on Significant Findings……………….. 36

LIST OF FIGURES

1. The biopsychosocial model of psychopathology …………………………………….… 31

2. Percent frequency for each of the 17 different types of SLEs …………………………. 32

3. Mean Y-BOCS total scores across temporal SLEs groups …………………………….. 33

4. Mean SI-R total score across temporal SLEs groups ………………………………….. 34

5. Mean number of total comorbid psychiatric disorders in addition to OCD across ……. 35 temporal SLEs groups.

LIST OF ABBREVIATIONS

OCD Obsessive Compulsive Disorder SLE Stressful Life Events CSA Childhood Sexual Abuse PTSD Post Traumatic Stress Disorder CRF Corticotropin Releasing Factor HPA Hypothalamic Pituitary Adrenal ACTH Adrenocorticotropic Hormone NIMH National Institute of Mental Health DSM-IV Diagnostic and Statistical Manual of Mental Disorders-Fourth Edition SCID-P Structured Clinical Interview for the Diagnostic and Statistical Manual of Mental Disorders-Patient Edition Y-BOCS Yale-Brown Obsessive Compulsive Scale & Checklist BDI Beck Depression Inventory SI-R Savings Inventory-Revised

ABSTRACT

Numerous studies have shown that throughout the human lifespan, stressful events influence the onset and course of mood and anxiety disorders. With regard to obsessive- compulsive disorder (OCD), life-stress has been implicated as a factor in the onset of the disorder, as well as a trigger for symptom relapse. These investigations not withstanding, relatively few empirical studies have investigated the relationship between traumatic stressful life events and OCD within a pathoplastic vulnerability framework. The present study builds on the current literature by examining the potential relationships between stress and obsessive- compulsive symptomatology, including symptom severity, symptom dimensions, and psychological comorbidity. Participants included 200 outpatients from the Adult OCD Clinic at the National Institute of Mental Health. Data are generally supportive of a pathoplastic relationship between SLEs and OCD symptomatology. It was demonstrated that SLEs are linked with increased OCD symptom severity and increased mood and anxiety disorder comorbidity. Additional analyses revealed that SLEs are more strongly associated with specific symptom dimensions, including obsessions/checking and symmetry/ordering. The hoarding symptom dimension, as measured by the Saving Inventory-Revised, also demonstrated a relationship to SLEs. Implications of these findings and future directions are discussed.

vii

INTRODUCTION

Anxiety disorders as a group have been identified as one of the most prevalent, chronic, and costly forms of psychopathology (Dupont, 1993, March; Kessler et al., 1994). In particular, obsessive compulsive disorder (OCD), an anxiety disorder manifested by distressing and recurrent obsessions and compulsions, has been named as one of the ten leading causes of disability worldwide (Lopez & Murray, 1998). Historically considered a rare disorder in the general population, recent epidemiological findings have indicated that OCD has a lifetime prevalence of approximately 2- 3% (e.g. Karno, Golding, Sorenson, & Burnam, 1988; Weissman et al., 1994). While research in the last few decades has expanded our understanding of the classification and treatment of OCD, many of the mechanisms and vulnerability factors involved in the etiology and maintenance of this disorder remain unknown. One putative vulnerability factor for psychiatric disorders is stressful life events (SLEs). Despite mounting evidence that suggests aversive experiences may play a role in the development and course of mood and anxiety disorders, the function of SLEs in OCD has been only marginally investigated. The purpose of this proposal is to briefly review the extant literature regarding the role of SLEs in the etiology and maintenance of anxiety disorders, with special emphasis on OCD. This investigation will subsequently examine the potential relationship(s) between SLEs and OCD within the context of a pathoplastic vulnerability model. Stressful Life Events and Anxiety Disorders In his book Hemmung, Symptom, und Angst, Freud (1926) theorized that anxiety was a reaction to experiencing events involving danger and loss. Since these early writings, the notion that traumatic or aversive life events cause mental disorders has received increasing empirical attention. Numerous studies have shown that extremely upsetting experiences, such as SLEs involving threat, defeat, humiliation or loss, influence the onset and course of psychological disorders throughout the human lifespan (e.g., Kendler, Neale, Kessler, Heath, & Eaves, 1993). In their comprehensive review of the literature, Paykel and Dowlatshahi (1988) concluded that SLEs were associated with a number of disorders including anxiety and depression. It appears that SLEs often precede episodes of anxiety (e.g., Angst & Vollrath, 1991) as well as mood

disorders (e.g., Bifulco, Brown, & Adler, 1991). There is also some indication that loss events are often associated with depression, while danger events are linked with anxiety (Finlay-Jones & Brown, 1981), though there is some debate over this distinction (e.g., Faravelli, 1985). Evidence derived from investigations of diathesis-stress theories equally suggests that SLEs may act as catalysts for the onset of symptoms (e.g., Brown, 1998; Caspi et al., 2003; Venturello, Barzega, Maina, & Bogetto, 2002). Aversive life experiences have also been associated with specific anxiety disorders. Horesh and colleagues (1997) found that events of loss and uncontrollable SLEs are linked with panic disorder, while additional investigations indicate that SLEs often precede the onset of panic disorder (e.g., Manfro et al., 1996). One study found that 80% of participants described one or more SLEs prior to their first panic attack (Roy-Byrne, Geraci, & Uhde, 1986). Another example was an investigation of a diathesis stress model that demonstrated that exposure to aversive conditions not only interacted with anxiety sensitivity to predict panic attacks, but also contributed to generalized emotional dysfunction (Zvolensky, Kotov, Antipova, & Schmidt, 2005). Individuals with generalized anxiety disorder (e.g., Newman & Bland, 1994) and social phobia (e.g., Brown, 1998) reportedly experience significantly more SLEs than normal controls. In addition, participants with generalized anxiety disorder perceive these events as more stressful and adapt to them less well than non-disordered individuals (e.g., Blazer, Hughes, & George, 1987; Newman & Bland, 1994). While the literature reviewed above does appear to provide support for the association between SLEs and anxiety disorders, these investigations represent correlational analyses. A stronger argument for causality emerges from analogue laboratory studies that promulgate the idea that environmental stress leads to distress. For example, Zvolensky et al. (2001) found that an individual’s perception of stress predicted fear responses to a biological challenge. Similar results were found in an investigation with panic disordered patients (Sanderson, Rapee, & Barlow, 1989). This study demonstrated that the perception of control ultimately influenced panic reaction to the biological challenge. Those individuals who did not have perceived control and were therefore in the more stressful condition, reported a greater number of anxiety symptoms and increased subjective anxiety. Taken as a whole, these findings suggest that the perception of stress, or rather the subjective experience of life events perceived as stressful, is a key factor in the processes involved in anxiety related responding.

Stressful Life Events and OCD Although the impact of SLEs on the development and maintenance of psychiatric disorders has been extensively examined, the direct association between SLEs and OCD has received relatively less empirical attention. One line of research addressing this relationship is a series of experimental investigations that have largely relied on non-clinical samples. These studies have assessed the function of generally-stressful experiences in generating non-specific anxiety and OC symptoms. Horowitz and colleagues conducted a series of experiments in which OCD patients and non-patients were exposed to stressful and aversive material (e.g., films). It was found that in response to the stressful-stimuli, participants reported an increase in intrusive thoughts (Horowitz, 1975; Horowitz & Becker, 1971). Rachman and de Silva (1978) came to similar conclusions using a large non-clinical sample. Their investigation revealed that external cues play an important role in triggering specific obsessional thoughts. A replication study by Horowitz et al. (1975) confirmed this finding, underscoring the tremendous impact the environment can have in inducing obsessive impulses. A more recent investigation conducted by Jones and Menzies (1998) found that in situations deemed stressful and dangerous, undergraduate subjects reported significantly higher ratings of anxiety, as well as a heightened urge to act out their obsessions/impulses. A further study has shown that high thought suppression coupled with low perceived control over stressful events was associated with increased OC symptomatology (McLaren & Crowe, 2003). Although these investigations have primarily dealt with non-clinical OC symptoms and generally stressful experiences (as opposed to traumatic SLEs), this evidence has significant clinical implications. As Rachman noted, intrusive thoughts “are after all the raw material for full obsessions” (Rachman, 1997, p. 797). Pierre Janet, an early anxiety disorder theorist, believed that OCD was caused by extreme emotional shock (Janet, 1903; Pitman, 1987). As evidence for his theory, Janet referenced the case of a woman who developed OCD after seeing the dead body of her daughter who had tragically perished in a house fire. This example does not appear to be an anomaly. Numerous modern case studies highlight the potential role that extreme stress and trauma play in inducing OC symptoms (e.g., Lipper & Feigenbaum, 1976; Rajarethinam, Abelson, & Himle, 2000; Trumbull, 2001). As previously reviewed, while experimental investigations appear to support a general association between life-stress and OCD, these case examples suggest that a unique relationship between traumatic negative life events and OCD may also exist.

The association between severe negative life-stress and OCD has been underscored through investigations of traumatized populations. Saunders and colleagues (1992) assessed a community sample of adult women for histories of childhood sexual abuse (CSA) and lifetime prevalence of mental disorders. They found that compared to non-victims, individuals who had experienced CSA were approximately five times more likely to have OCD. This association can also be seen in other types of populations who have experienced trauma. An investigation of combat-exposed Israeli veterans showed increased OC symptom severity (sub-clinical) when compared to normal controls (Solomon, 1993). Similarly, Jordan and colleagues (1991) found that the current prevalence of OCD in high-war-zone exposed Vietnam War veterans was 5.2%, which was significantly higher than that found in low-war-zone exposed veterans (0.3%). The link between trauma and OCD is also evident in samples of individuals with comorbid post-traumatic stress disorder (PTSD). Pitman and others have described case studies that clearly tie the onset of OCD, in addition to the onset of PTSD, to an extremely upsetting event such as high-combat exposure, industrial accidents, or sexual assault (de Silva & Marks, 1999; Pitman, 1993). Taken as a whole, these case studies appear to provide support for a traumatic origin model of OCD. Using a general population epidemiological survey, one of the few empirical investigations to address the association between OCD and PTSD found that the risk for OCD is 10 times greater in PTSD versus other psychiatric populations (Helzer, Robins, & McEvoy, 1987). While the exact relationship between symptoms of OCD and PTSD is currently unknown, investigations such as this seemingly support an association between severe SLEs and OCD. Stressful Life Events and the onset of OCD As noted above, the function of life-stress in OCD has been only marginally investigated. Studies of clinical populations have provided partial support for a direct association between SLEs and OCD. One of the earliest studies was conducted by Pollitt et al. (1957), and found that 62% of patients in an “obsessional state” believed that a specific event led to the onset of symptoms. Through case observations, it was shown that in many instances the illness was highly reactive to stressful environmental factors. Ingram (1961) and Lo (1967) attempted to extend and replicate these findings. By analyzing case histories in conjunction with diagnostic interviews, these investigations found that 64-69% of patients experienced a SLE six months to a year prior to the onset of the disorder. While these studies do appear to implicate SLE in the

onset of “obsessional neurosis,” conclusions drawn from their findings are limited in their scope due to methodological shortcomings. For instance, the results of these investigations are confounded due to the following: different premorbid study periods were operationalized, stressful/traumatic event criteria were not well defined, and consistent diagnoses were not employed. A recent series of investigations have attempted to further assess the relationship between SLEs and OCD onset. McKeon and colleagues (1984) assessed life events in 25 individuals with OCD using Paykel’s Life Events Schedule. (Of note, the SLEs included in this schedule are not necessarily negative or traumatic.) It was found that compared to non-clinical controls, OCD subjects experienced a greater number of SLEs in the year preceding onset. By contrast, an investigation of a 136 individuals (68 with OCD and 68 healthy controls) using similar methodology, found no such difference (Maina, Albert, Bogetto, Vaschetto, & Ravizza, 1999). The authors subsequently examined the specific types of SLEs and discovered that female OCD participants were more likely to have been pregnant than control participants. The investigators concluded that the post-partum period, and perhaps the stress associated with that period, represents a risk factor for OCD. Other investigations examining the link between pregnancy (as a SLE) and OCD, have been less conclusive (e.g., Williams & Koran, 1997). Gothelf and colleagues (2004) found that pediatric OCD patients reported significantly more total life events and increased negative life events in the year preceding OCD onset when compared to healthy controls. Finally, Hartl et al. (2005) found that compulsive hoarders (N=26; 32% with a reported diagnosis of OCD) related a significantly greater number of different types of trauma and more frequent traumatic experiences when compared to controls. Measurements were obtained via the Traumatic Events Scale-lifetime, which assesses 16 different traumatic events according to frequency and severity (Gershuny, Baer, Jenike, Minichiello, & Wilhelm, 2002). Interestingly, hoarding and non-hoarding groups did not differ in reported severity of disturbing events, nor in the degree of fear, helplessness, or horror they experienced. Due to the lack of standardized diagnostic methods, as well as the focus on hoarders, the generalizability of this investigation is limited. As a whole, the investigations reviewed above do appear to provide partial support for the theory that OCD onset is concomitant with generally SLEs. Inter-study agreement is inconclusive though, and only a minority of investigations has assessed the association between trauma-related negative life events and OCD onset.

Stressful Life Events and the course of OCD The relationship between SLEs and the course of OCD has been examined in very few investigations. Although the original studies conducted by Lo (1967), Ingram (1961), and Pollitt (1957) briefly examined the prognostic value that SLEs may or may not have, the general focus was on onset. Moreover, the association with course was not systematically assessed. More recently, Rasmussen and Eisen (1991) found that OCD patients reported a worsening of OC symptoms after experiencing generally SLEs. In an investigation designed to develop the Yale Children’s Global Stress Index, it was found that children with Tourette’s syndrome and/or OCD experienced significantly more psychosocial stress when compared to normal controls (Findley et al., 2003). The authors concluded that children with Tourette’s syndrome and/or OCD appear to experience a higher number of SLEs, which may play a role in the waxing and waning of symptoms often seen in pediatric OCD populations. In summary, the literature addressing the relationship between SLEs and OCD does seem to provide a certain degree of support that implicates traumatic life-stress as a factor in the onset and maintenance of OCD. These investigations not withstanding, the exact relationship between SLEs and OCD remains an empirical question, particularly with regard to traumatic negative life events. The majority of studies that have investigated the association between SLEs and OCD have been limited by (1) small sample sizes, (2) the difficulty of establishing retrospectively the temporal relationship between onset and SLEs (e.g., definition of premorbid period), and (3) a limited scope with regard to the effect of SLEs on OCD (most investigations have concentrated on symptom severity only). In order for definitive conclusions to be drawn, it will be necessary to address some of these limitations. SLEs that Occur in Childhood vs. Adulthood The relationship between SLEs and OCD may also be considered in light of when in an individual’s lifespan the SLE occurred. Mounting evidence suggests that early life-stress, in particular, may preferentially incline individuals to develop adult psychiatric disorders (Arborelius, Owens, Plotsky, & Nemeroff, 1999; Strohle & Holsboer, 2003). For example, early- life adversity such as CSA has been associated with a variety of problems including depression and anxiety (e.g., Brown & Harris, 1993; Kessler, 1997; Mullen, Martin, Anderson, Romans, & Herbison, 1996). In a large epidemiological investigation, McCauley and colleagues (1997) found that childhood abuse (independent of adult abuse) was related to a range of physical and

psychosocial difficulties. A prospective study of adolescents, which addressed the specificity of childhood stressful-experiences in predicting mood versus anxiety disorders, found similar results (Phillips, Hammen, Brennan, Najman, & Bor, 2005). Not only were childhood adversities associated with mood and anxiety disorders, but the occurrence of anxiety disorders, in particular, was significantly predicted (in relation to mood disorders). This finding held true even when maternal psychological disorders and current stressors were controlled. In summary, the compounded results from these investigations clearly indicate that adverse experiences in childhood and adolescence significantly elevate the risk for depression and anxiety conditions in adulthood. This association between childhood experiences of severe life-stress and increased adult psychopathology may be explained by the profound effect early-life experiences have on the developing brain. Nemeroff and colleagues have amassed considerable evidence indicating that changes in the developing brain can lead to life-long psychiatric sequelae (e.g., Nemeroff, 2004). Preclinical and clinical investigations conducted by Nemeroff and others have implicated several neurobiological systems, such as the corticotropin releasing factor (CRF) system, in the etiology of mood and anxiety disorders. In conjunction with the hypothalamic-pituitary-adrenal (HPA) axis, this system represents the major mammalian neuroendocrine stress response system (Kaufman, Plotsky, Nemeroff, & Charney, 2000). Hypothalamic CRF is released by an organism in response to stress that leads to the stimulation of the HPA axis, which in turn results in the secretion of the adrenocorticotropic hormone (ACTH). ACTH stimulates the secretion of other neuropeptides, which ultimately releases cortisol (Dallman et al., 1994). Through a number of well-executed analogue (mainly rat and non-human primates) and human studies (e.g., Coplan et al., 1996; Heim & Nemeroff, 2001), Nemeroff and colleagues have proposed a neurochemical hypothesis that would explain how early life-stress antecedes psychological disorders, such as anxiety disorders. It is hypothesized that particularly early life-stress leads to neuronal changes that result in the induction of persistently-elevated neuronal releases of CRF. The net effect is an increased responsiveness to stress (Nemeroff, 2004). Several investigations have shown that this increased responsiveness then renders individuals more susceptible to psychopathology in adulthood (Heim & Nemeroff, 2002).

Models of Vulnerability Factors in Psychopathology Modern understanding of the way psychiatric diseases manifest themselves is frequently grounded in a biopsychosocial framework (see Figure 1). This framework suggests a multilevel phenomenon and assumes that a disease entity functions within and across the different levels. With regard to OCD, genetics and neurological abnormalities likely are at play in the biological level. On the other hand, the psychological level includes a number of cognitive deficits, such as the level of importance a person places on intrusive thoughts. Finally, as an example of a social level component, researchers have identified markedly impaired social skills, which in turn lead to interpersonal stress and increased OC symptoms. All three levels will most likely be represented in an individual’s overall OCD symptomatology. Moreover, these levels will likely interact with one another. A prime example of this intra-level interaction can be seen in a diathesis-stress model where environmental forces act on a pre-existing vulnerability to trigger the disease state. The vast majority of investigations assessing the association between SLEs and OCD have been grounded within such a model. It is also important to consider a number of different vulnerability models, in addition to the diathesis-stress model, that provide a viable alternative explanation for the relationship between alleged vulnerability factors (e.g., SLEs) and the disorder. Klein, Wonderlich, and Shea (1993) and others (Clark, Watson, & Mineka, 1994) have discussed four different vulnerability models that fit within the larger biopsychosocial framework. The first model is the predisposition model, which presupposes that the vulnerability factor plays a causal role in the development of the disorder. The scar or complication model suggests that the disorder itself affects the vulnerability factor. Thirdly, the spectrum or continuity model states that the vulnerability factor and the disorder reflect the same construct. Finally, the pathoplastic model presumes that the vulnerability factor modifies the course and/or expression of the disorder. These models are not mutually exclusive and it is plausible that the relationship between a vulnerability factor and a given disorder could fit a number of models. In addition to the more traditional vulnerability model (diathesis-stress/predisposition model), the pathoplastic model may be a useful means of describing the association between SLEs and OCD. According to this model, individual differences in the experience of SLEs could effect the expression of the clinical condition. This is regardless of the role that stress may or may not play in the development of OCD. Rather, this specific model purports that SLEs modify

the manifestation and/or course of OCD after onset. As far as could be determined, there have been no explicit attempts to assess a pathoplastic relationship between SLEs and OCD. Present Study Aims and Hypotheses Is it then—as Janet theorized (1903) —that OCD is influenced by SLEs, such as extreme emotional shock? Based on the literature reviewed, this remains an empirical question. The present study seeks to examine the pathoplastic relationship between severe and/or traumatic life-stress and obsessive-compulsive symptomatology. This investigation differs from previous research in that it will be a more comprehensive analysis of this relationship and will explicitly examine the pathoplastic model within a large adult clinical sample of OCD. Accordingly, the first aim of this study will be to conduct a more systematic analysis of the association between SLEs and a variety of clinical facets of OCD. Consistent with previous research, it is hypothesized that SLEs will exacerbate the clinical expression of OCD resulting in a more complicated psychiatric phenomenon. It is further hypothesized that the experience of SLEs will be related to increased OC symptom severity. In the assessment of a purported vulnerability factor such as SLEs, one needs to address the potential spuriousness of that variable. Specifically, it will be important to demonstrate that SLEs are uniquely associated with OCD and that they are not better accounted for by a third variable, such as depression (e.g., Schmidt, Lerew, & Joiner, 1998). We furthermore hypothesize that certain OC symptom dimensions, such as hoarding and obsessions/checking, will be more strongly associated with SLEs than the other symptom dimensions (contamination/washing or symmetry/ordering). Compared to individuals with non- hoarding symptoms, individuals who demonstrate hoarding obsessions and compulsions have been shown to put a substantial emphasis on safety and frequently use their possessions as a source of comfort and safety (Sartory, Master, & Rachman, 1989; Hartl et al., 2005). Based on these findings as well as several case-reports, it has been posited that fear about potential harm and focus on safety may result from hoarding individuals having experienced more traumatic SLEs (e.g., Frost & Hartl, 1996) than non-hoarding individuals. In the same vein, it is possible that the obsessions/checking dimension (with the emphasis on security and certainty) will evidence a stronger relationship to SLEs than either the symmetry/ordering or contamination/washing dimensions.

A third hypothesis is that SLE will be more highly associated with amplified psychological comorbidity. Increased comorbidity is plausible considering that SLEs have been consistently associated with a range of anxiety disorders. In addition, the CSA literature suggests that traumatic events may lead to an extremely complicated clinical picture with increased psychological comorbidity (e.g., Felitti, 1991; Saunders et al., 1992). Comorbidity in OCD has been associated with poorer treatment response and is considered a poor prognostic indicator (Carrasco, Hollander, Schneier, & Liebowitz, 1992; Steketee, Eisen, Dyck, Warshaw, & Rasmussen, 1999; Stewart et al., 2004). As such, this issue of elevated psychiatric comorbidity has important treatment and course implications. The second aim of this investigation is to assess the influence that the temporal nature of an individual’s SLE has on his or her psychopathology. Based on the literature reviewed, it is plausible that those participants who encountered an adverse experience in childhood will have increased psychosocial difficulties when compared to individuals who experienced a SLE in adulthood. Moreover, those participants who experienced SLE in childhood and adulthood, will likely be more severe than those who experienced an event only in childhood or adulthood. It is hypothesized that childhood SLEs will be related to increased psychological comorbidity, as well as more severe OC symptomatology.

METHOD1

Participants The sample consisted of 265 outpatients from the Adult OCD Clinic at the National Institute of Mental Health (NIMH). These patients were part of a number of investigational and treatment protocols and were recruited through physician/psychologist referrals, NIMH websites, information booths at OCD and other psychological conferences, and advertisements in local newspapers. Inclusion criteria for participation included being at least 18 years of age and having an OCD diagnosis based on the Structured Clinical Interview for DSM-IV (SCID-P; First, 2001). Exclusion criteria included active schizophrenia or psychosis, severe mental retardation that does not permit an evaluation to characterize OCD, or OCD symptoms that occur exclusively in the context of depression. Measures Structured Clinical Interview for DSM-IV (SCID). Participants were interviewed using the SCID-P for DSM-IV (First, 2001), which is a semistructured interview designed to determine lifetime and current diagnoses of OCD and other major DSM Axis I disorders. This instrument consists of an initial overview section, followed by nine modules designed to assess specific domains of current and past psychological symptoms. The SCID interviews were administered by trained and clinically experienced interviewers, in either an in-person or telephone interview. In an effort to reach further reliability and interviewer consensus, each participant's SCID was re- assessed by two independent clinicians for blind diagnoses. Segal, Hersen, and Van Hasselt (1994) found the SCID was highly reliable for most Axis I disorders. The SCID has further been shown to have a kappa of 0.61 for 21 Axis I disorders (Williams & Koran, 1997). SCID diagnoses in previous studies conducted by this group have also demonstrated excellent reliability. Yale-Brown Obsessive Compulsive Scale (Y-BOCS). The magnitude of obsessions and compulsions was measured with the Y-BOCS (Goodman et al., 1989b), a frequently used assessment tool for the severity of obsessive-compulsive symptoms. The Y-BOCS is a global measure of functioning and total Y-BOCS scores have been shown to have high interrater

reliability (r = .72–.98, p < .05–.0001) (Goodman et al., 1989b; Richter, Cox, & Direnfeld, 1994). Y-BOCS ratings are significantly correlated with the modified version of Clinical Global Impression Scale for Global Severity of Obsessive–Compulsive Disorder (CGI-OCS; r = .74, p < .0001) (Goodman et al., 1989a; Guy, 1976). Specific symptoms were assessed using the Y- BOCS Symptom Checklist (Y-BOCS-SC) an obsessive-compulsive symptom inventory that has demonstrated high reliability and validity (e.g., Leckman et al., 1997). The Y-BOCS-SC was used to derive factor scores for the symptom dimensions of OCD. Specifically, principal component analysis was applied to the 13 apriori Y-BOCS categories and initial factor solutions were then rotated to simple structure with the Varimax procedure (e.g., Leckman et al., 1997). The factor analysis generated four factor scores for each subject representing the correlation of the symptom profile of the subject with each factor. The factor scores were standardized, with mean = 0 and standard deviation = 1 (Hasler et al., 2005). These symptom dimensions have been shown to be temporally stable and are associated with differential treatment response (Mataix- Cols, Rosario-Campos, & Leckman, 2005) and comorbidity patterns (Hasler et al., 2005). Stressful Life Events Measures. Life-stress was assessed via diagnostic interview. Specifically, the information provided in the “Traumatic Events List” contained in the PTSD module of the SCID was used to calculate four different measures of SLEs. The PTSD section of the SCID starts by providing participants with a brief description of extremely upsetting, stressful or traumatic life experiences. Following this concise definition of a severe SLE, the participants are asked an open-ended question regarding their own experience of such events. The SCID interviewer then proceeds to record all experiences the participant reports, as well as the age when the event occurred. The participant is then asked if they are still troubled by any of the incidents they recounted. If so, the interviewer proceeds with the official PTSD module. This method assures that regardless of whether or not an individual meets criteria for PTSD, a list of SLEs has been documented. This list is the subjective report of each participant, according what they deem stressful or traumatic in their lives. It is highly unlikely that the establishment of SLEs using this methodology would produce an overestimate of such life events. Rather, the possibility of an underestimate of SLEs is more probable. The four specific SLEs measures are as follows: (1) Stressful/Traumatic Life Events. This is a dichotomous measure that assesses whether or not an individual has ever experienced a SLE in his or her lifetime.

(2) Number of Stressful/Traumatic Life Events. Each event that an individual experiences will be assigned one count. The sum of all events experienced will serve as a continuous measure of SLE exposure. (3) Weighted Stress Measure. The PTSD literature (e.g., Chang, Connor, Lai, Lee, & Davidson, 2005; Roy-Byrne et al., 2004) has documented that while some life events are considered universally stressful (e.g., loosing a loved one), others put individuals at an increased risk for developing PTSD (e.g., criminal victimization). By assigning each stressful event listed in the PTSD section to either a "PTSD" category or a "high stress" category, a continuous measure of stress can be created that takes into account event severity. Those events under the "PTSD" category will be given 2 points, while those in the "high stress" category will be assigned 1 point. The total score will be the sum of all weighted events the individual has experienced. (4) Temporal Experiences of Stressful/ Traumatic Life Events. Depending on when in an individual’s life-span the stressful or traumatic event occurred, the participants will be placed into four categories: (a) No SLEs, if the individuals has never experienced a traumatic SLE; (b) Childhood SLEs, if an event occurred prior to age 18; (c) Adulthood SLEs, if an event occurred after age 18; or (d) Childhood & Adulthood SLEs, if the individual experienced events both childhood and adulthood. The age cut-offs are consistent with other studies that compared the effects of childhood vs. adult abuse (e.g., McCauley et al., 1997). Beck Depression Inventory (BDI). The BDI is a 21-item measure of global depression that assesses current mood. It is a reliable and validated measure of depressive symptomatology (e.g., Beck, Epstein, Brown, & Steer, 1988). Saving Inventory-Revised (SI-R). The SI-R (Frost, Steketee, & Grisham, 2004) is a 23- item questionnaire designed to measure hoarding behaviors. Questions require participants to define the extent to which a symptom is present on a scale from 0 to 4; the anchors differ depending on the item. This instrument includes three subscales for clutter (9 items), difficulty discarding (7 items) and acquisition (7 items). The SI-R has been found to have strong internal consistency (Coles, Frost, Heimberg, & Steketee, 2003), good test-retest reliability, and good convergent validity (Frost et al., 2004).

Data Analytic Plan The investigation conducted by Hartl and colleagues (2005) contained a SLEs measure most consistent with that proposed for this investigation and may therefore serve as the most appropriate comparison study. The effect sizes found by Hartl et al. were moderate (d’s = 0.45- 0.59). Using a statistical power analysis program (Faul & Erdfelder, 1992), it was determined that 130 participants would allow examination of the present study hypotheses at a power > 80% (β < .20) to test a moderate effect size with a Type 1 error (α) < 0.05. The present sample should be adequate to test the main study hypotheses. In accordance with the first aim, which was to examine the relationship between SLEs and a variety of clinical facets of OCD, a series of regression analyses were utilized. The analyses for each specific hypothesis are outlined below. SLEs and OCD symptom severity. For the hypothesis that SLEs are associated with increased OC symptom severity, linear regression analyses were used to determine if SLEs predicted increased Y-BOCS scores. In addition, a covariance strategy, as outlined by Kendall and Ingram (1989), was used to assess the predictive specificity of SLEs. Two regression equations were constructed to assess the potential non-spuriousness of SLEs. For the first equation, with Y-BOCS total score as the dependent variable, BDI scores were entered into the equation followed by SLEs. For the second equation, BDI scores served as the dependent variable, while Y-BOCS total scores were entered as the covariate. This approach allowed for the assessment of the relation of SLEs to OCD symptoms beyond the effects of depression symptoms, as well as the assessment of the relation of SLEs to depression beyond the effects of OCD symptoms. SLEs and OCD symptom dimensions. The association between SLE and symptom domains was evaluated in a number of ways. Zero-order correlations among these measures were evaluated along with a test of correlated correlations (Meng, Rosenthal, & Rubin, 1992). In addition, linear regression analyses were used to evaluate the unique variance accounted for by each OC symptom dimension on SLE. Specifically, all four of the symptom dimension scores were regressed onto the SLEs measure to determine which symptom dimension contributed the most variance. SLEs and psychological comorbidity. For the hypothesis that SLEs are associated with amplified psychological comorbidity, two types of regression analyses were performed. Logistic

regression analyses were used to determine if SLEs predict a dichotomized variable of psychological comorbidity: (1) comorbidity: the individual has comorbid disorders in addition to OCD, or (2) none: the individual has no comorbid disorders. As part of a more comprehensive investigation of this hypothesis, linear regression analyses were used to determine if SLEs predicted increased psychological comorbidity as measured by (1) the total number of comorbid disorders in addition to OCD, (2) the total number of comorbid mood disorders, and (3) the total number of comorbid anxiety disorders. Effects of Timing of SLE and Symptomatology. The second aim, which was to examine the effect of the temporal experience of a SLE, was investigated utilizing ANOVA. Specifically, data were analyzed across the dependent variable of interest and between groups (No SLEs; Childhood SLEs; Adulthood SLEs; Childhood & Adulthood SLEs) using one-way ANOVA with Tukey’s WSD post-hoc comparison. The dependent variables of interest included OC symptom severity as measured by the Y-BOCS and the SI-R, and psychological comorbidity as measured by (1) the total number of comorbid disorders in addition to OCD, (2) the total number of comorbid mood disorders, and (3) the total number of comorbid anxiety disorders.

RESULTS2

Zero-order correlations among the key variables, in addition to means, standard deviations, and alpha internal consistency coefficients (when applicable) are presented in Table 1. It was found that 54% of the sample experienced at least one SLE and that the maximum number of SLEs experienced by any one person was five. Of those individuals who had experienced a SLE, 42.2% (N = 51) experienced SLEs in adulthood only, 42.2% (N = 46) in childhood only, and 16.4% (N = 19) experienced SLEs in both adulthood and childhood. Of those events experienced (N=238), accident participant was the most frequent, with 16% of events falling into this category. The next four most frequent events were: nonviolent death witnessed (10.1%), followed by crime witnessed (9.2%), sexual abuse (9.2%), and illness of self or loved one (8.0%). Figure 2 depicts a bar-graph of the frequency percentages for all seventeen SLE categories. Due to no clear consensus in the literature, as well as the novelty of our SLEs measures, it was unclear how to best operationalize SLEs. This methodological issue was explored by simultaneously regressing all three SLEs measures onto Y-BOCS total score. It was found that the dichotomous measure made the strongest unique contribution to OCD symptom severity (beta = .2), followed by total number of SLEs (beta = -.016) and the weighted measure (beta = .009). In addition to this analysis, the significant findings using each of the three measures were compared across all subsequent analyses (see Appendix A, Table 1). In general, results using all three SLEs indicators were fairly consistent, though the dichotomous SLEs measure appeared to be the clearest and provided the most consistent findings. Based on these initial analyses, only the results using the dichotomous SLEs measure will be presented below. Our first hypothesis was that SLEs would be associated with increased OC symptom severity and that this relationship would be non-spurious. Detailed results for all analyses are presented in Table 2. For each dependent variable a series of equations were constructed (denoted A-D) to accommodate different combinations of covariates in addition to the primary independent variable (SLE). Linear regression analyses revealed that SLEs predicted increased Y-BOCS total scores (β = .19, t(241)= 3.06, p = .001), in addition to increased Y-BOCS obsessions (β=.17, t(241)= 2.6, p=.001) and compulsions (β = .17, t(241)= 2.74, p = .001) scores.

Analyses were repeated with age and age-of-onset as covariates. These variables were chosen because age-of-onset is associated with increased Y-BOCS symptom severity and age is associated with the experience of SLEs (the longer an individual lives, the more likely they are to have experienced a SLE). Results indicated that despite controlling for these covariates, SLEs significantly predicted Y-BOCS total score, Y-BOCS obsession score, and Y-BOCS compulsion score (see Table 2). In order to evaluate if SLEs have a unique association with OCD, we conducted separate follow-up analyses with depression as an additional covariate. Results indicated that SLEs were a significant predictor of Y-BOCS total score beyond the effects of age, age-of-onset, and BDI scores (β = .16, t(159)= 1.97, p = .049). In contrast, the SLEs prediction of BDI scores controlling for age, age-of-onset, and Y-BOCS scores was a non-significant trend (β = .15, t(159)= 1.79, p = .08). These findings signify that SLEs are specifically related to OCD but not to depression. To test our second hypothesis, that certain symptom dimensions would be more strongly associated with SLEs than others, we were required to compare the four OCD symptom dimensions to one another. As such, we first calculated alpha internal consistency coefficients to assure that the four Y-BOCS symptom dimension measures (using the apriori Y-BOCS categories to represent each factor), as well as the supplemental SI-R hoarding measure, had comparable reliabilities. The alpha coefficients (see Table 1) were fairly consistent across the symptom dimensions and were deemed acceptable (greater than 0.7). One exception was the alpha coefficient for the SI-R, which was exceptional. Next, we calculated zero-order correlations between SLEs and the symptom dimension factor scores. It was found that Y-BOCS obsessions/checking was significantly and positively correlated with SLEs (r(242) = .22, p = .001). Y-BOCS symmetry/ordering (r(242) = .17, p = .01) and SI-R hoarding (r(178) = .25, p = .001) were also significantly and positively correlated with the experience of a SLE. In contrast, neither Y-BOCS contamination/cleaning (r(242) = .06, p = .35) nor the Y-BOCS hoarding dimension (r(242) = .07, p = .25) were significantly correlated with SLEs. We then proceeded to compare these correlations to one another in an attempt to identify those factors that were most strongly associated with SLEs. Following the procedure outlined by Meng and colleagues (1992), we found no significant differences between any of the correlations. We also examined the unique variance accounted for by each OCD symptom dimension with regard to SLEs, by

simultaneously regressing all four symptom dimensions onto the SLEs measure. Because the SI- R had a much higher reliability than the Y-BOCS hoarding measure (alpha coefficients = 0.97 versus 0.71), we decided to conduct separate analyses for each. Accordingly, we constructed two logistic regression analyses: (1) using only the Y-BOCS symptom dimensions, and (2) substituting the SI-R for the Y-BOCS hoarding symptom dimension. It was found that in both equations obsessions/checking explained the most variance, followed by symmetry/ordering, then hoarding (whether measured with the SI-R or the Y-BOCS), and finally contamination/cleaning (see Table 3). To further investigate the association between the symptom dimensions of OCD and SLEs, a series of additional exploratory linear hierarchical regression analyses were conducted. Similar to the symptom severity analyses discussed above, we wanted to determine if SLEs would predict symptom dimension scores even when controlling for covariates (age, age-of- onset, and BDI scores). These analyses were conducted for all four of the Y-BOCS factors, as well as the SI-R and its three subscales (clutter, difficulty discarding, and acquisitioning). Results are presented in Table 4. It was found that SLEs significantly predicted scores on obsessions/checking (β = .22, t(242)= 3.49, p = .001), symmetry/ordering (β = .17, t(242)= 2.68, p = .01), hoarding SI-R total (β = .25, t(178)= 3.35, p = .001), and all three of the SI-R subscales (see Table 4). When age, age-of-onset, and BDI scores were added as covariates to the regression equation, the prediction of obsessions/checking (β = .16, t(242)= 2.32, p = .02) and symmetry/ordering (β = .17, t(242)= 1.79, p = .03) remained significant. Hoarding SI-R total (β = .16, t(178)= 1.94, p = .06) and the SI-R clutter subscale (β = .15, t(178)= 1.77, p = .08) were non-significant trends. The final hypothesis for Aim 1 was that SLEs would be associated with increased comorbidity. We used a series of linear regression analyses to investigate this hypothesis with the various comorbidity measures (total number of comorbid disorders, anxiety disorders, and mood disorders) as the dependent variable, SLEs as the independent variable, and age and age- of-onset as covariates. Of note, individuals with a comorbid PTSD diagnosis were removed from the sample for these analyses only, so that they would not confound the results. Findings are presented in Table 5. Although we had originally planned on testing a dichotomous measure of comorbidity, the finding that 92% of the sample had at least one comorbid disorder led us to conclude that such an analysis would not be informative. With only 8% of the sample having no

comorbidity, a dichotomous indicator would have greatly reduced variability and as such, the chances of finding meaningful differences would have been slim. Results using the three remaining comorbidity variables revealed that SLEs did not predict the total number of comorbid disorders, though, SLEs were associated with a greater number of additional anxiety disorders (β = .14, t(218)= 2.09, p = .04) and greater number of comorbid mood disorders (β = .15, t(219)= 2.11, p = .04) after controlling for age and age-of-onset. A series of one-way ANOVAs were conducted to investigate the effect of SLE occurrence (No SLEs; Childhood SLEs; Adulthood SLEs; Childhood & Adulthood SLEs) on OCD symptom severity as measured by the Y-BOCS total and the SI-R (hoarding symptom severity). Results revealed a significant group effect for Y-BOCS total (F = 3.329; df = 3, 224; Levene test, p = .956; means test p = .02). Using Tukey’s HSD, however, the group comparisons revealed no statistically significant differences. Figure 3 depicts the mean Y-BOCS scores for each of the four groups. The SI-R also showed a significant main effect (F = 3.414; df = 3, 162; Levene test, p = .145; means test p = .019). Group comparisons for this dependent variable revealed a non-significant trend with those individuals who experienced a SLE in childhood having greater SI-R scores than those who experienced no SLE (see Figure 4 for a graphical depiction of mean SI-R scores for each group). A related hypothesis was that the four categories of participants would differ in levels of comorbidity. Analyses revealed a significant group effect for total number of comorbid psychological disorders (F = 3.95; df = 3, 216; Levene test, p = .064; means test p = .009). Comparisons between the groups revealed that those who had experienced SLE in childhood had significantly more comorbid psychiatric disorders than those individuals who had never experienced a SLE (see Figure 5 for a graphical depiction of mean total comorbidity for each group). Next, we found a significant group effect for total number of comorbid anxiety disorders (F=7.091; df=3, 227; Levene test, p = .000; means test p = .001) and total number of mood disorders (F=7.956; df=3, 230; Levene test, p = .000; means test p = .001). In both cases, those individuals who experienced SLEs in adulthood and childhood only had greater levels of mood and anxiety comorbidity, compared to those who had never experienced a SLE. For mood disorders, it was also found that those who experienced SLEs in both childhood and adulthood had significantly greater levels of comorbidity than those who reported no SLEs. There were no significant differences between any of the three SLEs groups on any of the dependent variables.

DISCUSSION

The current study investigated the potential pathoplastic relationship between SLEs and OCD symptomatology across a number of clinical domains. Data are generally supportive of such a relationship. Consistent with our hypothesis that SLEs would exacerbate the clinical expression of OCD, results revealed that there was a relatively robust relationship between experiencing a SLE and increased OC symptom severity. This finding corroborates previous investigations, such as the experimental studies on non-clinical samples (McLaren & Crowe, 2003) and the handful of clinical investigations that have assessed the impact of SLE on OCD severity (Rasmussen & Eisen, 1991). Importantly, our study demonstrated that this relationship remained significant despite controlling for a number of influential factors, particularly depressive symptomatology. To our knowledge, this is the first investigation that has assessed the non-spuriousness of the association between SLEs and OCD. The covariance strategy employed demonstrated that SLEs predicted OC symptom severity beyond the effects of depressive symptoms. In contrast, SLEs did not predict BDI scores controlling for OC symptom severity. We furthermore predicted that the experience of SLEs would be more strongly associated with specific OC symptom dimensions (hoarding and obsessions/checking). Our hypothesis was partially supported. SLEs were significantly correlated with the obsessions/checking and symmetry/ordering symptom dimensions, as well as the SI-R hoarding dimensions. The strengths of these correlations, however, were not significantly different from one another. Additional analyses revealed that the obsessions/checking dimension explained the most variance within the SLEs measure, followed by symmetry/ordering, hoarding (whether measured via the SI-R or the Y-BOCS), and finally contamination cleaning. Based on these initial findings, we conducted a series of follow-up regression analyses to determine if SLEs predicted symptom dimension scores, while covarying age, age-of-onset, and depressive symptomatology. Data were consistent with the results from the previous analyses in that higher scores on both the obsessions/checking and symmetry/ordering dimensions had a robust association with SLEs. SI-

R total scores and the SI-R clutter subscale also demonstrated a fairly robust relationship with SLEs; however these results emerged as non-significant trends. Upon initial examination, it may appear surprising that the two measures of the hoarding symptom dimension had such divergent results—a fairly consistent relationship between SLEs and the SI-R, with no significant association between SLEs and the Y-BOCS hoarding dimension. Two possible factors may have contributed to this discrepancy. First, the SI-R had a much higher alpha coefficient compared to the Y-BOCS hoarding dimension, which provided the SI-R with considerable greater reliability. This enhanced reliability is not surprising given that the SI-R has 23 items (versus the two hoarding items of the Y-BOCS), and most likely afforded us with better power to detect differences. A second factor that may have influenced the lack of agreement between the two measures of hoarding is that the two items of the Y-BOCS, unlike the SI-R, do not tap into the three theoretically and empirically grounded components of hoarding—clutter, difficulty discarding, and acquisitioning (Grisham & Barlow, 2005). The follow-up regression analyses revealed that the clutter subscale had the most consistent relationship with SLEs. Based on this information, it is plausible that the divergent findings between the Y-BOCS and the SI-R occurred not only because of discrepant reliability of the measures, but also because the clutter symptoms are not assessed in the Y-BOCS; the Y-BOCS primarily taps acquisitioning, which demonstrated the weakest association with SLEs. Consistent with our third hypothesis that SLEs would be associated with amplified comorbidity, our results demonstrated a relationship between the experience of a traumatic event and a greater number of comorbid mood and anxiety disorders. These findings converge with the substantial literature addressing the association between SLEs and mood/anxiety disorders, particularly major depression (e.g., Bifulco, Brown, & Adler, 1991; Manfro et al., 1996). We found that the total number of comorbid disorders was not significantly associated with SLEs. This may suggest that SLEs are more closely related to mood and/or anxiety disorder comorbidity, compared to other comorbid disorders (e.g., Tourette’s disorder, trichotillomania, substance abuse). Although informative, our investigation is unable to speak to causality. It is likely that multiple pathways account for these comorbidity findings. One possibility is that SLEs lead, independently of OCD, to the development of mood and other anxiety disorders. Alternatively, it may be that SLEs act through a common vulnerability pathway that leads to a comorbidity phenotype of OCD. A third possibility is that OCD, when combined with SLEs, acts

as a trigger for the expression of comorbid disorders. One important piece of information for answering these questions would be to determine if the relationship between SLEs and comorbidity in OCD is limited to, or stronger with, specific mood and anxiety disorders. Because comorbidity is so prevalent in samples of OCD (92% of participants received one or more additional Axis I diagnoses) and because it is tied to increased symptom severity (LaSalle et al., 2004) and poorer treatment response (Carrasco et al., 1992), clarifying comorbidity patterns has significant treatment implications. Moreover, demonstrating that certain disorders occur more frequently in individuals with OCD and their relatives can provide compelling evidence for potential etiological relationships (Pauls et al., 1995; Nestadt et al., 2003). Comorbidity may also help explain our findings that the obsessions/checking and symmetry/ordering symptom dimensions had the strongest association with SLEs. A recent investigation by Hasler et al. (2005) found that obsessions/checking and symmetry/ordering are especially associated with mood and anxiety disorders. Obsessions/checking was associated with comorbid major depression, dysthymia, generalized anxiety disorder, panic disorder, social and specific phobias, whereas symmetry/ordering was associated with comorbid bipolar disorders and panic disorder. Neither hoarding, nor the contamination/cleaning symptom dimensions were strongly associated with mood and/or anxiety disorders. Considering these findings in light of the extensive and robust evidence tying mood and other anxiety disorders to SLEs (Brown & Harris, 1993), a model emerges that may explain our results. Perhaps obsessions/checking and symmetry/ordering are associated with SLEs through their increased comorbidity with mood and anxiety disorders. Future studies should assess the link between SLEs and specific obsessions/compulsions taking into consideration comorbidity patterns. Again, such investigations could elucidate potential etiological relationships and pathways. We hypothesized that those individuals who experienced SLEs in both childhood and adulthood would have greater symptom severity and comorbidity than those who experienced SLEs in childhood or adulthood only. With regard to the effect of the temporal experience of a SLE, results showed that childhood SLEs were significantly different in hoarding symptom severity, but not general OC symptom severity, compared to those who had never experienced an SLE. Also, those individuals who experienced SLEs in either childhood or adulthood differed significantly from those who reported no SLEs in mood and anxiety disorder comorbidity. Overall, these results only partially confirm our hypothesis. Considering that the combined

childhood and adulthood SLEs group had only 19 members, it is likely that we lacked power to detect differences. Nevertheless, our findings partially corroborate the evidence gathered from neurobiological investigations, such as hyperactivation of the CRF system and the HPA axis (Nemeroff, 2004). To our knowledge, this is the first study to have examined this hypothesis in a sample of individuals with OCD. Future studies are warranted given that we found differences between the groups despite our small sample size and retrospective methodology. Taken as a whole, results from our study provide initial support for a pathoplastic relationship between SLEs and OCD symptomatology. We extended the extant literature by assessing this relationship in a large fully-evaluated OCD sample and across a number of domains including symptom severity, symptom dimensions, and comorbidity. Moreover, analyses demonstrated that the association between SLEs and OCD is non-spurious, at least in this sample. Further investigation of the role that SLEs play in OCD is vital considering that, no environmental hypothesis has been presently proposed with strong empirical support (July 5, 2005 correspondence, OCGS). Although we have demonstrated a pathoplastic relationship, we are unable to rule out the possibility that the relationship between OCD and SLEs would better fit a predisposition model. Garber and Hollon (1991) outlined three criteria for demonstrating that a certain variable, such as SLEs, acts as a vulnerability factor. The following must be demonstrated: (1) covariance between the vulnerability factor and an outcome; (2) a non- spurious relationship between the vulnerability factor and an outcome; and (3) the temporal precedence of a vulnerability factor. Due to our methodological constraints we are unable to assess the third criterion specified. We have, however, met the first two requirements. Our findings should be considered in light of several limitations, such as the retrospective collection of data. Numerous researchers have discussed the problems inherent in this design, particularly when examining the influence of SLEs on psychopathology (Maina, 1999). Acknowledging this limitation we were careful not to draw any conclusions about the relationship between SLEs and the onset of OCD. We also constrained ourselves to an investigation of a pathoplastic model. Our measures of SLEs are an additional constraint since they were based on the traumatic events listed in the PTSD SCID module. While this module has been thoroughly validated and consistently and reliably used to identify individuals with PTSD retrospectively, indices of reliability and validity for our specific measure of SLEs have not yet been established. Finally, the conclusions we were able to draw regarding the specificity

analyses, as well as those regarding comorbidity, were limited in that our sample was not necessarily a “pure” OCD sample—over 80% had a comorbid mood disorder. That being said, this rate of comorbidity is in no way an anomaly. Other investigations have found a similar prevalence of comorbid disorders, and it is considered very common for patients with OCD to be diagnosed with other psychiatric disorders (LaSalle et al., 2004). Thus, while the high degree of comorbidity may be representative of typical OCD samples, it also makes it extremely difficult to detect any potential differences. The fact that some differences did emerge provides credence to future investigations. The results generated from this study have multiple implications for future research in OCD. One avenue for future investigations would be to clarify the link between SLEs, OCD and comorbidity, particularly that with major depressive disorder. In addition, the distinction between a pathoplastic model and a predisposition model could be explored using experimental procedures utilizing laboratory stressors. Furthermore, it would be informative to conduct prospective clinical investigations to demonstrate the temporal association between SLEs and OCD. Regardless of whether or not SLEs are identified as a vulnerability factor for OCD, assessing the interaction between stress and a diathesis, such as specific gene susceptibility regions, could unearth potential pathogenic mechanisms.

FOOTNOTES

1The present report represents an archival study, and as such leads to certain restrictions and methodological limitations. These limitations not withstanding, this investigation would still make a vital contribution to the OCD literature.

2Additional exploratory analyses were suggested during the prospectus defense. To streamline the results section, these analyses will be presented in Appendix B, rather than in the body of the text.

Table 1 Intercorrelations Between OCD Variables and SLE Variable 1 2 3 4 5 6 7 8 9 10 11 M SD α 1 SLE -- .32 .19 .17 .17 .19 .22 .17 .06 .07 .25 na na na 2 PTSD Dx -- .15 .14 .13 .18 .22 .16 .17 .07 .21 na na na 3 Y-BOCS Total -- .91 .89 .28 .43 .37 .39 .10 .22 21.2 9.0 na 4 Y-BOCS Obsess -- .62 .28 .48 .33 .30 .05 .14 10.8 5.3 na 5 Y-BOCS Comp -- .21 .28 .34 .40 .13 .24 10.5 4.8 na 6 BDI -- .49 .20 .33 .22 .31 14.6 10.9 na 7 Factor 1 -- .44 .40 .22 .21 na na 0.71 8 Factor 2 -- .40 .23 .04 na na 0.73 9 Factor 3 -- .11 .20 na na 0.75 10 Factor 4 -- .60 na na 0.71 11 SI-R -- 27.0 19.5 0.97

Note. 1, SLE = dichotomous measure of stressful life event experienced; 2, PTSD Dx = Post Traumatic Stress Disorder diagnosis; 3, Y-BOCS Total = Yale-Brown Obsessive-Compulsive Scale total score; 4, Y-BOCS Obsess = Yale-Brown Obsessive-Compulsive Scale obsessions score; 5, Y-BOCS Comp = Yale-Brown Obsessive-Compulsive Scale compulsions score; 6, BDI = Beck Depression Inventory; 7, Factor 1 = Y-BOCS obsessions/checking; 8, Factor 2 = Y-BOCS symmetry/ordering; 9, Factor 3 = Y-BOCS contamination/cleaning; 10, Factor 4 = Y-BOCS hoarding; 11, SI-R = Saving Inventory-Revised. na = not applicable

Table 2 Summary of Hierarchical Regression Analyses for SLE Prediciting OCD Symptom Severity 2 DV Equation Step Predictor R B SE B β t Y-BOCS A 1 SLE 0.04 3.50 1.14 0.19 3.06 ** total 1 age 0.04 -0.03 0.04 -0.05 -0.67 B age of onset -0.18 0.07 -0.19 -2.70 ** 2 SLE 0.04 3.65 1.16 0.20 3.16 ** 1 BDI 0.08 0.24 0.07 0.29 3.64 ** C 2 SLE 0.02 2.62 1.43 0.15 1.84 trend 1 age 0.04 -0.03 0.05 -0.05 -0.53 age of onset -0.18 0.08 -0.19 -2.17 * D 2 BDI 0.07 0.22 0.07 0.27 3.37 ** 3 SLE 0.02 2.83 1.44 0.16 1.97 *

Y-BOCS A 1 SLE 0.03 1.74 0.67 0.17 2.60 ** obsess 1 age 0.05 -0.04 0.02 -0.12 -1.73 B age of onset -0.08 0.04 -0.15 -2.22 * 2 SLE 0.03 1.94 0.67 0.18 2.87 ** 1 BDI 0.08 0.14 0.04 0.28 3.59 ** C 2 SLE 0.01 1.22 0.84 0.12 1.47 1 age 0.05 -0.04 0.03 -0.12 -1.39 age of onset -0.08 0.05 -0.15 -1.77 D 2 BDI 0.07 0.13 0.04 0.26 3.34 ** 3 SLE 0.02 1.46 0.84 0.14 1.74 trend

Y-BOCS A 1 SLE 0.03 1.67 0.61 0.17 2.74 ** comp 1 age 0.03 0.01 0.02 0.03 0.45 B age of onset -0.09 0.04 -0.18 -2.60 ** 2 SLE 0.03 1.65 0.62 0.17 2.64 ** 1 BDI 0.04 0.09 0.04 0.21 2.57 ** C 2 SLE 0.02 1.35 0.78 0.14 1.74 trend 1 age 0.03 0.01 0.03 0.03 0.36 age of onset -0.09 0.04 -0.18 -2.08 * D 2 BDI 0.04 0.08 0.04 0.19 2.34 * 3 SLE 0.02 1.34 0.79 0.14 1.71 trend

BDI A 1 SLE 0.04 4.14 1.71 0.19 2.42 * 1 age 0.01 -0.01 0.06 -0.02 -0.18 B age of onset -0.13 0.10 -0.11 -1.25 2 SLE 0.04 4.25 1.78 0.19 2.38 * 1 Y-BOCS 0.08 0.35 0.10 0.29 3.64 ** C 2 SLE 0.02 3.04 1.73 0.14 1.76 trend 1 age 0.01 -0.01 0.06 -0.02 -0.18 age of onset -0.13 0.10 -0.11 -1.25 D 2 Y-BOCS 0.07 0.33 0.10 0.27 3.37 ** 3 SLE 0.02 3.18 1.78 0.15 1.79 trend Note. SLE = stressful life event experienced yes/no; Y-BOCS = Yale-Brown Obsessive-Compulsive Scale total score; Y-BOCS Obsess = Yale-Brown Obsessive-Compulsive Scale obsessions score; Y- BOCS Comp = Yale-Brown Obsessive-Compulsive Scale compulsions score; BDI = Beck Depression Inventory; age = age at time of study; age of onset = age at which OCD onset * p < 0.05, ** p<0.01, trend = 0.1 > p > 0.05

Table 3 Comparison of the Four Symptom Dimensions

DV Predictor B Wald Exp(B) C.I. SLE SIR 0.03 6.91 ** 1.03 1.0-1.1 Factor 1 0.18 0.75 1.20 .8-1.8

Factor 2 0.17 0.69 1.18 .8-1.7 Factor 3 -0.13 0.48 0.87 .6-1.3

Factor 1 0.33 3.70 * 1.31 1.0-1.9 Factor 2 0.23 1.88 1.26 .9-1.7 Factor 3 -0.21 1.75 0.81 .6-1.1 Factor 4 0.01 0.01 1.01 .8-1.4

Note. SLE = dichotomous measure of stressful life event experienced; Factor 1 = Y-BOCS obsessions/checking; Factor 2 = Y-BOCS symmetry/ordering; Factor 3 = Y-BOCS contamination/cleaning; Factor 4 = Y-BOCS hoarding; SI-R = Saving Inventory-Revised. p < 0.05, ** p<0.01

Table 4 Summary of Hierarchical Regression Analyses for SLE Predicting OCD Symptom Dimensions 2 DV Equation Step Predictor R B SE B β t Factor 1 A 1 SLE 0.05 0.43 0.12 0.22 3.49 ** 1 age 0.11 -0.01 0.01 -0.17 -2.04 * age of onset -0.02 0.01 -0.24 -2.81 ** B 2 BDI 0.21 0.04 0.01 0.46 6.64 ** 3 SLE 0.03 0.32 0.14 0.16 2.32 *

Factor 2 A 1 SLE 0.03 0.34 0.13 0.17 2.68 * 1 age 0.08 -0.01 0.01 -0.20 -2.31 * age of onset -0.02 0.01 -0.15 -1.76 B 2 BDI 0.03 0.02 0.01 0.18 2.23 * 3 SLE 0.03 0.34 0.16 0.17 2.15 *

Factor 3 A 1 SLE 0.00 0.12 0.13 0.06 0.93 1 age 0.06 0.00 0.01 -0.04 -0.51 age of onset -0.02 0.01 -0.22 -2.55 ** B 2 BDI 0.09 0.03 0.01 0.30 3.91 ** 3 SLE 0.00 0.01 0.16 0.01 0.07

Factor 4 A 1 SLE 0.05 0.14 0.13 0.07 1.14 1 age 0.01 -0.01 0.01 -0.09 -1.00 age of onset -0.01 0.01 -0.05 -0.54 B 2 BDI 0.04 0.02 0.01 0.21 2.54 * 3 SLE 0.00 0.10 0.16 0.05 0.58

SI-R total A 1 SLE 0.06 9.56 2.85 0.25 3.35 ** 1 age 0.06 0.32 0.12 0.25 2.76 ** age of onset -0.31 0.19 -0.15 -1.69 B 2 BDI 0.09 0.55 0.15 0.31 3.76 ** 3 SLE 0.02 6.24 3.23 0.16 1.94 trend

SI-R A 1 SLE 0.06 2.16 0.89 0.18 2.42 * aquisitioning 1 age 0.00 0.18 2.72 0.01 0.07 age of onset -1.88 4.35 -0.04 -0.43 B 2 BDI 0.01 -4.85 3.53 -0.11 -1.36 3 SLE 0.00 26.30 79.95 0.03 0.33

SI-R A 1 SLE 0.06 4.08 1.28 0.24 3.20 ** clutter 1 age 0.05 0.14 0.05 0.25 2.62 ** age of onset -0.09 0.08 -0.10 -1.10 B 2 BDI 0.11 0.26 0.07 0.33 4.01 ** 3 SLE 0.02 2.55 1.44 0.15 1.77 trend

SI-R A 1 SLE 0.04 2.78 1.10 0.19 2.53 ** difficulty 1 age 0.05 0.11 0.05 0.23 2.40 * discarding age of onset -0.10 0.07 -0.14 -1.44 B 2 BDI 0.03 0.12 0.06 0.19 2.12 * 3 SLE 0.02 1.91 1.30 0.13 1.47 Note. SLE = dichotomous measure of stressful life event experienced; Factor 1 = Y-BOCS obsessions/checking; Factor 2 = Y-BOCS symmetry/ordering; Factor 3 = Y-BOCS contamination/cleaning; Factor 4 = Y-BOCS hoarding; SI-R = Saving Inventory-Revised. * p < 0.05, ** p<0.01, trend = 0.1 > p > 0.05

Table 5 Summary of Hierarchical Regression Analyses for SLE Predicting Comorbidity

DV Equation Step Predictor R2 B SE B β t Total A 1 SLE 0.01 0.29 0.22 0.09 1.32 Comorbidity 1 age 0.02 -0.01 0.01 -0.05 -0.62 B age of onset -0.02 0.01 -0.11 -1.37 2 SLE 0.01 0.32 0.23 0.10 1.39 Total A 1 SLE 0.02 0.31 0.13 0.16 2.32 * Anxiety Comorbidity 1 age 0.02 0.01 0.01 0.09 1.18 B age of onset -0.01 0.01 -0.13 -1.67 2 SLE 0.02 0.29 0.14 0.14 2.09 * Total Mood A 1 SLE 0.03 0.30 0.13 0.16 2.36 * Comorbidity 1 age 0.01 0.01 0.01 0.10 1.32 B age of onset -0.01 0.01 -0.11 -1.42 2 SLE 0.02 0.28 0.13 0.15 2.11 *

Note. Total Comorbidity = total number of comorbid disorders in addition to OCD; Total Anxiety Comorbidity = total number of comorbid anxiety disorders in addition to OCD; Total Mood Comorbidity = total number of comorbid mood disorders. Individuals with PTSD diagnosis were removed from the comorbidity analyses. * p < 0.05, ** p<0.01

Figure 1. The biopsychosocial model of psychopathology.

Precentage of Types of SLEs Experienced

accident participant

death witnessed, nonviolent

crime witnessed by self

sexual abuse

illness of loved one or self

physical abuse

other

rape

death witnessed, violent

abuse NOS Event Type attempted rape

crime witnessed by other

technological disaster

natural disaster

accident witnessed

miliatry duty

abandonment

0.0% 2.0% 4.0% 6.0% 8.0% 10.0% 12.0% 14.0% 16.0% 18.0% Percentage

Figure 2. Percent frequency for each of the 17 different types of SLEs.

Mean Ybocs total Ybocs Mean 20

no SLE experienced SLE in adulthood SLE in childhood SLE in both adult & childhood None vs Adult vs Child vs Both SLE

Figure 3. Mean Y-BOCS total scores across temporal SLEs groups. Note. Error bars = +/- 2 standard errors

35 *

30 Mean SI-R Mean

no SLE experienced SLE in adulthood SLE in childhood SLE in both adult & childhood None vs Adult vs Child vs Both SLE

Figure 4. Mean SI-R total score across temporal SLEs groups. Note. Error bars = +/- 2 standard errors * non-significant trend difference between SLE in childhood and no SLE experienced, p = .052

4.5

4.0

3.5 **

3.0

2.5

2.0 Mean Total Number of Comorbid Disorders Comorbid of NumberTotal Mean

1.5

no SLE experienced SLE in adulthood SLE in childhood SLE in both adult & childhood None vs Adult vs Child vs Both SLE

Figure 5. Mean number of total comorbid psychiatric disorders in addition to OCD across temporal SLEs groups. Note. Error bars = +/- 2 standard errors ** significant difference between SLE in childhood and no SLE experienced, p = .03

APPENDIX A Table 6 Comparison of the three SLE measures based on significant findings.

SLE TotalSLE SLEweighted Controlling for: Controlling for: Controlling for: SLE age & age, onset, SLE age & age, onset, SLE age & age, onset, onset BDI BDI onset BDI BDI onset BDI BDI Y-BOCS x x trend x Y-BOCS x x Y-BOCS x x Y-BOCS Y-BOCS Y-BOCS obsess obsess obsess x x trend x Y-BOCS Y-BOCS Y-BOCS comp x x trend trend comp x x comp x x

Factor 1 x x trend x Factor 1 x x Factor 1 x x Factor 2 x x x Factor 2 x x Factor 2 x x Factor 3 Factor 3 Factor 3 Factor 4 Factor 4 Factor 4

SI-R total x x x trend SI-R total x x x trend SI-R total x x x x SI-R SI-R SI-R aquisitioning x x aquisitioning aquisitioning SI-R clutter x x x trend SI-R clutter x x x trend SI-R clutter x x x x SI-R SI-R SI-R difficulty x x difficulty x trend difficulty trend

Total Total Total comorbidity comorbidity comorbidity Total anxiety Total anxiety Total anxiety comorbidity x x dx x x dx x x Total mood Total mood Total mood comorbidity x x dx x x dx x x Note. SLE = dichotomous stressful life event measure; Y-BOCS Total = Yale-Brown Obsessive-Compulsive Scale total score; Y-BOCS Obsess = Yale-Brown Obsessive- Compulsive Scale obsessions score; Y-BOCS Comp = Yale-Brown Obsessive-Compulsive Scale compulsions score; Factor 1 = aggressive, sexual, religious and somatic obsessions, and checking compulsions as assessed by the Y-BOCS; Factor 2 = obsessions of symmetry, and repeating, counting and ordering compulsions; Factor 3 = contamination obsessions and cleaning compulsions; Factor 4 = hoarding obsessions and compulsions; SI-R = Saving Inventory-Revised, aquisitioning, clutter, and difficulty to discard subscales; Total comorbidity = total number of comorbid conditions in addition to OCD; Total anxiety comorbidity = total number of comorbid anxiety disorders; Total mood comorbidity = total number of comorbid mood disorders. x =significance at the p<0.05 level, trend = 0.1 > p > 0.05

APPENDIX B

An additional exploratory analysis was suggested during the prospectus defense. There is some evidence to suggest that early-onset OCD is a phenotypically distinct group compared to late-onset OCD

(Tukel et al., 2005). One investigation assessed a range of neuropsychological functions in late- versus

early-onset OCD. The authors concluded that diverging neurobiological mechanisms seemed to be

expressed in the two groups (Roth et al., 2005). One explanation for these findings is that it appears as if

early-onset OCD represents a more heritable phenotype. For example, early-onset has been found to be

related to an increased family history of OCD, as well as comorbidity with Tourrette’s Syndrome and

specific genetic variants in the dopamine receptor type 4 gene (Delorme et al., 2005; Hemmings et al.,

2005). We hypothesized that age-of-onset would interact with the experience of a SLE, such that SLEs

would have less of an effect on individuals with early-onset OCD. Hierarchical regression analyses were

conducted to assess not only the main effects of onset and SLEs (see Table 2), but also the interaction

between age-of-onset and SLEs. The results from these analyses are depicted below for each of the three

SLEs measures. No significant interaction effects were found.

2 DV Step Predictor R B SE B β t 1 age of 0.08 -0.19 0.06 -0.20 -3.20 ** Y-BOCS onset SLE 3.45 1.13 0.19 3.06 ** SLE 2 SLExOnset 0.01 -0.14 0.12 -0.16 -1.11

1 age of 0.06 -0.19 0.06 -0.20 -3.21 ** Y-BOCS onset SLE 1.14 0.51 0.14 2.25 * 2 SLExOnset 0.00 -0.04 0.05 -0.11 -0.87 Total SLE

1 age of 0.06 -0.19 0.06 -0.20 -3.20 ** Y-BOCS onset SLE 0.63 0.29 0.14 2.17 * 2 SLExOnset 0.00 -0.02 0.03 -0.06 -0.49

SLE weighted

Note. Y-BOCS = Yale-Brown Obsessive-Compulsive Scale total score; age of onset = age at which OCD onset; SLE = Stressful life events measure (SLE = has individual ever experienced a stressful life event; TotalSLE = total number of stressful life events experienced; SLEweighted = total number of events taking into account severity of event); SLExOnset = interaction between onset and stressful life event measure. * p < 0.05, ** p<0.01

REFERENCES

Angst, J., & Vollrath, M. (1991). The natural history of anxiety disorders. Acta psychiatrica Scandinavica, 84(5), 446-452.

Arborelius, L., Owens, M. J., Plotsky, P. M., & Nemeroff, C. B. (1999). The role of corticotropin-releasing factor in depression and anxiety disorders. The Journal of endocrinology, 160(1), 1-12.

Beck, A. T., Epstein, N., Brown, G., & Steer, R. A. (1988). An inventory for measuring clinical anxiety: Psychometric properties. Journal of Consulting and Clinical Psychology, 56(6), 893-897.

Bifulco, A., Brown, G. W., & Adler, Z. (1991). Early sexual abuse and clinical depression in adult life. British Journal of Psychiatry, 159, 115-122.

Blazer, D., Hughes, D., & George, L. K. (1987). Stressful life events and the onset of a generalized anxiety syndrome. American Journal of Psychiatry, 144(9), 1178-1183.

Brown, G. W. (1998). Genetic and population perspectives on life events and depression. Social psychiatry and psychiatric epidemiology, 33(8), 363-372.

Brown, G. W., & Harris, T. O. (1993). Aetiology of anxiety and depressive disorders in an inner- city population. 1. Early adversity. Psychological Medicine, 23(1), 143-154.

Carrasco, J. L., Hollander, E., Schneier, F. R., & Liebowitz, M. R. (1992). Treatment outcome of obsessive compulsive disorder with comorbid social phobia. Journal of Clinical Psychiatry, 53(11), 387-391.

Caspi, A., Sugden, K., Moffitt, T. E., Taylor, A., Craig, I. W., Harrington, H., et al. (2003). Influence of life stress on depression: Moderation by a polymorphism in the 5-htt gene. Science, 301(5631), 386-389.

Chang, C. M., Connor, K. M., Lai, T. J., Lee, L. C., & Davidson, J. R. (2005). Predictors of posttraumatic outcomes following the 1999 Taiwan earthquake. The Journal of Nervous and Mental Diseases, 193(1), 40-46.

Clark, L. A., Watson, D., & Mineka, S. (1994). Temperament, personality, and the mood and anxiety disorders. Journal of Abnormal Psychology, 103(1), 103-116.

Coles, M. E., Frost, R. O., Heimberg, R. G., & Steketee, G. (2003). Hoarding behaviors in a large college sample. Behaviour Research and Therapy, 41(2), 179-194.

Coplan, J. D., Andrews, M. W., Rosenblum, L. A., Owens, M. J., Friedman, S., Gorman, J. M., et al. (1996). Persistent elevations of cerebrospinal fluid concentrations of corticotropin- releasing factor in adult nonhuman primates exposed to early-life stressors: Implications for the pathophysiology of mood and anxiety disorders. Proceedings of the National Academy of Sciences of the United States of America, 93(4), 1619-1623.

Dallman, M. F., Akana, S. F., Bradbury, M. J., Strack, A. M., Hanson, E. S., & Scribner, K. A. (1994). Regulation of ACTH secretion. Seminars in Neuroscience, 6, 205-213. de Silva, P., & Marks, M. (1999). The role of traumatic experiences in the genesis of obsessive- compulsive disorder. Behaviour Research and Therapy, 37(10), 941-951.

Dupont, R. L. (1993, March). "Town hall" panel presentation. Paper presented at the The Anxiety Disorders of America Conference, Charleston, SC.

Faravelli, C. (1985). Life events preceding the onset of panic disorder. Journal of Affective Disorders, 9(1), 103-105.

Felitti, V. J. (1991). Long-term medical consequences of incest, rape, and molestation. The Southern medical journal, 84(3), 328-331.

Findley, D. B., Leckman, J. F., Katsovich, L., Lin, H., Zhang, H., Grantz, H., et al. (2003). Development of the Yale Children's Global Stress Index (YCGSI) and its application in children and adolescents with tourette's syndrome and obsessive-compulsive disorder. Journal of the American Academy of Child and Adolescent Psychiatry, 42(4), 450-457.

Finlay-Jones, R., & Brown, G. W. (1981). Types of stressful life event and the onset of anxiety and depressive disorders. Psychological Medicine, 11(4), 803-815.

First, M. B., Spitzer, R. L., Gibbon, M., & Williams, J. B. W. (2001). Structured clinical interview for dsm-iv-tr axis i disorders, research version, patient edition (scid-i/p).New York: Biometrics Research, New York State Psychiatric Institute.

Freud, S. (1926). Hemmung, symptom und angst. Frankfurt: Fischer Taschenbuecher. Frost, R.O. & Hartl, T.L. (1996). A cognitive-behavioral model of compulsive hoarding. Behaviour Research and Therapy, 34,341-350.

Frost, R. O., Steketee, G., & Grisham, J. (2004). Measurement of compulsive hoarding: Saving inventory-revised. Behaviour Research and Therapy, 42(10), 1163-1182.

Garber, J. & Hollon, S.D. (1991). What can specificity designs say about causality in psychopathology research? Psychological Bulletin, 110, 129-136.

Gershuny, B. S., Baer, L., Jenike, M. A., Minichiello, W. E., & Wilhelm, S. (2002). Comorbid posttraumatic stress disorder: Impact on treatment outcome for obsessive-compulsive disorder. American Journal of Psychiatry, 159(5), 852-854.

Goodman, W. K., Price, L. H., Rasmussen, S. A., Mazure, C., Delgado, P., Heninger, G. R., et al. (1989a). The Yale-Brown Obsessive Compulsive Scale. Ii. Validity. Archives of General Psychiatry, 46(11), 1012-1016.

Goodman, W. K., Price, L. H., Rasmussen, S. A., Mazure, C., Fleischmann, R. L., Hill, C. L., et al. (1989b). The yale-brown obsessive compulsive scale. I. Development, use, and reliability. Archives of General Psychiatry, 46(11), 1006-1011.

Gothelf, D., Aharonovsky, O., Horesh, N., Carty, T., & Apter, A. (2004). Life events and personality factors in children and adolescents with obsessive-compulsive disorder and other anxiety disorders. Comprehensive Psychiatry, 45(3), 192-198.

Grisham, J.R. & Barlow, D.H. (2005). Compulsive Hoarding: Current Research and Theory. Journal of Psychopathology and Behavioral Assessment, 27, (1), 45-52.

Guy, W. (1976). Ecdeu, assessment manual for psychopharmacology.Washington, DC: US Department of Health, Education, and Welfare.

Hartl, T. L., Duffany, S. R., Allen, G. J., Steketee, G., & Frost, R. O. (2005). Relationships among compulsive hoarding, trauma, and attention-deficit/hyperactivity disorder. Behaviour Research and Therapy, 43(2), 269-276.

Hasler, G., LaSalle-Ricci, V.H., Ronquillo, J.G., Crawley, S.A., Cochran, L.W., Kazuba, D., Greenberg, B.D., & Murphy, D.L. (2005). Obsessive-compulsive disorder symptom dimensions show specific relationships to psychiatric comorbidity. Psychiatry Research, 135, 121-132.

Heim, C., & Nemeroff, C. B. (2001). The role of childhood trauma in the neurobiology of mood and anxiety disorders: Preclinical and clinical studies. Biological Psychiatry, 49(12), 1023-1039.

Heim, C., & Nemeroff, C. B. (2002). Neurobiology of early life stress: Clinical studies. Seminars in clinical neuropsychiatry, 7(2), 147-159.

Helzer, J. E., Robins, L. N., & McEvoy, L. (1987). Post-traumatic stress disorder in the general population. Findings of the epidemiologic catchment area survey. New England Journal of Medicine, 317(26), 1630-1634.

Horesh, N., Amir, M., Kedem, P., Goldberger, Y., & Kotler, M. (1997). Life events in childhood, adolescence and adulthood and the relationship to panic disorder. Acta psychiatrica Scandinavica, 96(5), 373-378.

Horowitz, M. J. (1975). Intrusive and repetitive thoughts after experimental stress. A summary. Archives of General Psychiatry, 32(11), 1457-1463. Horowitz, M. J., & Becker, S. S. (1971). Cognitive response to stress and experimental demand. Journal of Abnormal Psychology, 78(1), 86-92.

Ingram, I. M. (1961). Obsessional illness in mental hospital patients. The Journal of mental science, 107, 382-402.

Janet, P. (1903). Les obsessions et la psychasthenie (2nd ed. Vol. 1 & 2). Paris: Alcan. Jones, M. K., & Menzies, R. G. (1998). Role of perceived danger in the mediation of obsessive- compulsive washing. Depression and anxiety, 8(3), 121-125.

Jordan, B. K., Schlenger, W. E., Hough, R., Kulka, R. A., Weiss, D., Fairbank, J. A., et al. (1991). Lifetime and current prevalence of specific psychiatric disorders among vietnam veterans and controls. Archives of General Psychiatry, 48(3), 207-215.

Karno, M., Golding, J. M., Sorenson, S. B., & Burnam, M. A. (1988). The epidemiology of obsessive-compulsive disorder in five us communities. Arch Gen Psychiatry, 45(12), 1094-1099.

Kaufman, J., Plotsky, P. M., Nemeroff, C. B., & Charney, D. S. (2000). Effects of early adverse experiences on brain structure and function: Clinical implications. Biological Psychiatry, 48(8), 778-790.

Kendall, P.C., & Ingram, R.E. (1989). Cognitive-behavioral perspectives: Theory and research on depression and anxiety. In P.C. Kendall & D. Watson (Eds.) Anxiety and depression: Distinctive and overlapping features. Personality, psychopathology, and psychotherapy (pp. 27-53). San Diego, CA: Academic Press.

Kendler, K. S., Neale, M., Kessler, R., Heath, A., & Eaves, L. (1993). A twin study of recent life events and difficulties. Archives of General Psychiatry, 50(10), 789-796.

Kessler, R. C. (1997). The effects of stressful life events on depression. Annual review of psychology, 48, 191-214.

Kessler, R. C., McGonagle, K. A., Zhao, S., Nelson, C. B., Hughes, M., Eshleman, S., et al. (1994). Lifetime and 12-month prevalence of dsm-iii-r psychiatric disorders in the united states. Results from the national comorbidity survey. Arch Gen Psychiatry, 51(1), 8-19.

Klein, M. H., Wonderlich, S., & Shea, M. T. (1993). Models of relationships between personality and depression: Toward a framework for theory and research. In M. H. Klein, S. Wonderlich & M. T. Shea (Eds.), Personality and depression: A current view.New York: Guilford Press.

LaSalle, V.H., Cromer, K.R., Nelson, K.N., Kazuba, D., Justement, L., Murphy, D.L., (2004). Diagnostic interview assessed neuropsychiatric disorder comorbidity in 334 individuals with obsessive-compulsive disorder. Depression and Anxiety, 19, 163-173. Leckman, J. F., Grice, D. E., Boardman, J., Zhang, H., Vitale, A., Bondi, C., et al. (1997). Symptoms of obsessive-compulsive disorder. American Journal of Psychiatry, 154(7), 911-917.

Lipper, S., & Feigenbaum, W. M. (1976). Obsessive-compulsive neurosis after viewing the fetus during therapeutic abortion. American Journal of Psychotherapy, 30(4), 666-674. Lo, W. H. (1967). A follow-up study of obsessional neurotics in hong kong chinese. British Journal of Psychiatry, 113(501), 823-832.

Lopez, A. D., & Murray, C. C. (1998). The global burden of disease, 1990-2020. Nat Med, 4(11), 1241-1243.

Mataix-Cols, D., Rosario-Campos, M.C., & Leckman, J.F.(2005). A multidimensional model of obsessive-compulsive disorder. American Journal of Psychiatry, 162 (2): 228-238.

Maina, G., Albert, U., Bogetto, F., Vaschetto, P., & Ravizza, L. (1999). Recent life events and obsessive-compulsive disorder (ocd): The role of pregnancy/delivery. Psychiatry Research, 89(1), 49-58.

Manfro, G. G., Otto, M. W., McArdle, E. T., Worthington, J. J., 3rd, Rosenbaum, J. F., & Pollack, M. H. (1996). Relationship of antecedent stressful life events to childhood and family history of anxiety and the course of panic disorder. Journal of Affective Disorders, 41(2), 135-139.

McCauley, J., Kern, D. E., Kolodner, K., Dill, L., Schroeder, A. F., DeChant, H. K., et al. (1997). Clinical characteristics of women with a history of childhood abuse: Unhealed wounds. Jama, 277(17), 1362-1368.

McKeon, J., Roa, B., & Mann, A. (1984). Life events and personality traits in obsessive- compulsive neurosis. British Journal of Psychiatry, 144, 185-189.

McLaren, S., & Crowe, S. F. (2003). The contribution of perceived control of stressful life events and thought suppression to the symptoms of obsessive-compulsive disorder in both non- clinical and clinical samples. Journal of Anxiety Disorders, 17(4), 389-403.

Meng, X., Rosenthal, R., & Rubin, D. B. (1992). Comparing correlated correlation coefficients. Psychological Bulletin, 111, 172-175.

Mullen, P. E., Martin, J. L., Anderson, J. C., Romans, S. E., & Herbison, G. P. (1996). The long- term impact of the physical, emotional, and sexual abuse of children: A community study. Child Abuse and Neglect, 20(1), 7-21.

Nemeroff, C. B. (2004). Neurobiological consequences of childhood trauma. Journal of Clinical Psychiatry, 65 Suppl 1, 18-28. Nestadt, G., Addington, A., Samuels, J., Liang, K.Y., Bienvenu, O.J., Riddle, M., et al., (2003). The identification of OCD-related subgroups based on comorbidity. Biological Psychiatry, 53, 914-920.

Newman, S. C., & Bland, R. C. (1994). Life events and the 1-year prevalence of major depressive episode, generalized anxiety disorder, and panic disorder in a community sample. Comprehensive Psychiatry, 35(1), 76-82.

Pauls, D.L., Alsobrook, J.P., Goodman, W., Rasmussen, S., Leckman, J.F. (1995). A family study of obsessive-compulsive disorder. American Journal of Psychiatry, 152, 76-84.

Paykel, E. S., & Dowlatshahi, D. (1988). Life events and mental disorder. In S. Fisher & J. Reason (Eds.), Handbook of life stress, cognition and health.New York: Wiley.

Phillips, N. K., Hammen, C. L., Brennan, P. A., Najman, J. M., & Bor, W. (2005). Early adversity and the prospective prediction of depressive and anxiety disorders in adolescents. Journal of Abnormal Child Psychology, 33(1), 13-24.

Pitman, R. K. (1987). Pierre Janet on obsessive-compulsive disorder (1903). Review and commentary. Archives of General Psychiatry, 44(3), 226-232.

Pitman, R. K. (1993). Posttraumatic obsessive-compulsive disorder: A case study. Comprehensive Psychiatry, 34(2), 102-107.

Pollitt, J. (1957). Natural history of obsessional states; a study of 150 cases. British Medical Journal, 32(5012), 194-198.

Rachman, S. (1997). A cognitive theory of obsessions. Behaviour Research and Therapy, 35(9), 793-802.

Rachman, S., & de Silva, P. (1978). Abnormal and normal obsessions. Behaviour Research and Therapy, 16(4), 233-248.

Rajarethinam, R. P., Abelson, J. L., & Himle, J. A. (2000). Acute onset and remission of obsessions and compulsions following medical illnesses and stress. Depression and anxiety, 12(4), 238-240.

Rasmussen, S. A., & Eisen, J. L. (1991). Phenomenology of ocd: Clinical subtypes, heterogeneity and coexistence. In J. Zohar, T. R. Insel & S. Rasmussen (Eds.), The psychobiology of obsessive-compulsive disorder.New York: Springer.

Richter, M. A., Cox, B. J., & Direnfeld, D. M. (1994). A comparison of three assessment instruments for obsessive-compulsive symptoms. Journal of behavior therapy and experimental psychiatry, 25(2), 143-147.

Roy-Byrne, P., Arguelles, L., Vitek, M. E., Goldberg, J., Keane, T. M., True, W. R., et al. (2004). Persistence and change of ptsd symptomatology--a longitudinal co-twin control analysis of the Vietnam era twin registry. Social psychiatry and psychiatric epidemiology, 39(9), 681-685.

Roy-Byrne, P. P., Geraci, M., & Uhde, T. W. (1986). Life events and the onset of panic disorder. American Journal of Psychiatry, 143(11), 1424-1427.

Sanderson, W. C., Rapee, R. M., & Barlow, D. H. (1989). The influence of an illusion of control on panic attacks induced via inhalation of 5.5% carbon dioxide-enriched air. Archives of General Psychiatry, 46(2), 157-162.

Sartory, G., Master, D., & Rachman, S.J. (1989). Safety-signal therapy in agoraphobics: A preliminary test. Behaviour Research and Therapy, 27, 205-210.

Saunders, B. E., Villeponteaux, L. A., Lipovsky, J. A., Kilpatrick, D. G., & Veronen, L. J. (1992). Child sexual assault as a risk factor for mental disorders among women. Journal of Interpersonal Violence, 7(2), 189-204.

Schmidt, N.B., Lerew, D.R., &Joiner, T.E (1998). Anxiety sensitivity and the pathogenesis of anxiety and depression: evidence for symptom specificity. Behaviour Research and Therapy, 36, 165-177.

Segal, D. L., Hersen, M., & Van Hasselt, V. B. (1994). Reliability of the structured clinical interview for dsm-ill-r: An evaluative review. Comprehensive Psychiatry, 35(4), 316- 327.

Slattery, M. J., Dubbert, B. K., Allen, A. J., Leonard, H. L., Swedo, S. E., & Gourley, M. F. (2004). Prevalence of obsessive-compulsive disorder in patients with systemic lupus erythematosus. Journal of Clinical Psychiatry, 65(3), 301-306.

Solomon, Z. (1993). Combat stress reactions: The enduring toll of war.New York: Plenum Press.

Steketee, G., Eisen, J., Dyck, I., Warshaw, M., & Rasmussen, S. (1999). Predictors of course in obsessive-compulsive disorder. Psychiatry Research, 89(3), 229-238.

Stewart, S. E., Geller, D. A., Jenike, M., Pauls, D., Shaw, D., Mullin, B., et al. (2004). Long-term outcome of pediatric obsessive-compulsive disorder: A meta-analysis and qualitative review of the literature. Acta psychiatrica Scandinavica, 110(1), 4-13.

Strohle, A., & Holsboer, F. (2003). Stress responsive neurohormones in depression and anxiety. Pharmacopsychiatry, 36 Suppl 3, S207-214.

Trumbull, D. W. (2001). Obsessive-compulsive symptomatology: A goal-directed response to anticipated traumatization? Psychiatry, 64(4), 309-318. Venturello, S., Barzega, G., Maina, G., & Bogetto, F. (2002). Premorbid conditions and precipitating events in early-onset panic disorder. Comprehensive Psychiatry, 43(1), 28- 36.

Weissman, M. M., Bland, R. C., Canino, G. J., Greenwald, S., Hwu, H. G., Lee, C. K., et al. (1994). The cross national epidemiology of obsessive compulsive disorder. The cross national collaborative group. J Clin Psychiatry, 55 Suppl, 5-10.

Williams, K. E., & Koran, L. M. (1997). Obsessive-compulsive disorder in pregnancy, the puerperium, and the premenstruum. Journal of Clinical Psychiatry, 58(7), 330-334; quiz 335-336.

Zvolensky, M. J., Eifert, G. H., & Lejuez, C. W. (2001). Offset control during recurrent 20% carbon dioxide-enriched air induction: Relation to individual difference variables. Emotion, 1(2), 148-165.

Zvolensky, M. J., Kotov, R., Antipova, A. V., & Schmidt, N. B. (2005). Diathesis stress model for panic-related distress: A test in a Russian epidemiological sample. Behaviour Research and Therapy, 43(4), 521-532.

BIOGRAPHICAL SKETCH

Kiara Cromer was born August 3, 1979. She attended Whitworth College in Spokane Washington, years 1998-2002, and earned her Bachelor of Science in Biochemistry. She entered the Department of Psychology at Florida State University in 2003 to pursue a doctoral degree in Clinical Psychology.