Journal of Affective Disorders 180 (2015) 116–121

Contents lists available at ScienceDirect

Journal of Affective Disorders journal homepage: www.elsevier.com/locate/jad

Research report

The relationship between rumination, PTSD, and depression symptoms Michelle E. Roley a, Meredith A. Claycomb a, Ateka A. Contractor a,b, Paula Dranger c, Cherie Armour d, Jon D. Elhai a,n a

University of Toledo, Department of Psychology, Toledo, OH, USA Department of Psychiatry & Human Behavior, Warren Alpert Medical School of Brown University, Providence, RI, USA c Choices! Counseling Services, Porter, LaPorte and Lake County, Valparaiso, IN, USA d Psychology Research Institute, Ulster University, Coleraine Campus, Coleraine, Northern Ireland, United Kingdom b

art ic l e i nf o

a b s t r a c t

Article history: Received 6 March 2015 Received in revised form 1 April 2015 Accepted 2 April 2015 Available online 9 April 2015

Background: Posttraumatic stress disorder (PTSD) and major depressive disorder (MDD) are highly comorbid (Elhai et al., 2008. J. Clin. Psychiatry, 69, (4), 597–602). Rumination is a cognitive mechanism found to exacerbate and maintain both PTSD and MDD (Elwood et al., 2009. Clin. Psychol. Rev. 29, (1), 87– 100; Olatunji et al., 2013. Clin. Psychol.: Sci. Pract. 20, (3), 225–257). Aims: Assess whether four rumination subtypes moderate the relationship between comorbid PTSD and MDD symptoms. Method: We consecutively sampled patients (N ¼45) presenting to a mental health clinic using selfreport measures of PTSD and MDD symptoms, and rumination in a cross-sectional design. Results: Repetitive rumination moderates the relationship between PTSD and MDD symptoms at one standard deviation above the mean (β ¼ .044, p¼ .016), while anticipatory rumination moderates the relationship between PTSD and MDD symptoms at mean levels and higher levels of anticipatory rumination (mean β ¼ .030, p ¼.042; higher β ¼.060, p ¼.008). Discussion: Repetitive and anticipatory rumination should be assessed in the context of comorbid PTSD and MDD and interventions should focus on reducing these rumination subtypes. Limitations: Results should be replicated with other trauma populations because the number and complexity of traumatic events may impact the assessed symptoms. Constructs should also be assessed longitudinally, in order to establish causality. We are unable to confirm why rumination styles moderated the relationship between PTSD and depression or why counterfactual thinking and problem-focused thinking did not moderate the relationship between the two constructs. & 2015 Elsevier B.V. All rights reserved.

Keywords: Depression PTSD Trauma Cognition Abuse/maltreatment/neglect

1. Introduction Posttraumatic stress disorder (PTSD) and major depressive disorder (MDD) are highly comorbid, with 48–55% of individuals with a lifetime diagnosis of PTSD also meeting the criteria for a diagnosis of MDD (Elhai et al., 2008). This high comorbidity rate persists even after controlling for overlapping symptoms (Elhai et al., 2008; Grubaugh, et al., 2010). Thus, it is important to investigate constructs that could play a role in the relationship between these highly comorbid diagnoses; our current study focuses on rumination. Rumination, the tendency to have repetitive, persistent, uncontrollable and intrusive thoughts (Brinker and Dozois, 2009; Rippere, 1977), is a cognitive mechanism found to exacerbate and maintain both PTSD

n Correspondence to: University of Toledo, Department of Psychology, 2801 W. Bancroft St. Toledo, Ohio 43606. Tel.: þ 1 419 530 2829: fax: þ 1 419 530 8479. E-mail address: [email protected] (J.D. Elhai).

http://dx.doi.org/10.1016/j.jad.2015.04.006 0165-0327/& 2015 Elsevier B.V. All rights reserved.

and MDD (see Elwood et al. (2009)) for review on rumination and PTSD; see Olatunji et al. (2013) for review on rumination and depression). Despite research on the role rumination plays separately in both PTSD and depression, little known research has examined the way rumination affects the relationship between the two disorders; particularly whether subtypes of rumination impact the relationship differently. 1.1. Rumination and PTSD Rumination is a maladaptive form of cognitive coping (Brewin and Holmes, 2003). Rumination is a distinct construct separate from PTSD theorized to maintain PTSD symptoms by cognitively avoiding thinking of traumatic memories (Ehlers and Clark, 2000). According to Ehlers and Clark's cognitive model of PTSD (Ehlers and Clark, 2000), which was reviewed in Brewin and Holmes (2003), rumination is a cognitive strategy that is utilized to reduce levels of distress that result from negative interpretations of intrusive traumatic memories. Therefore, rumination has been deemed a cognitive

M.E. Roley et al. / Journal of Affective Disorders 180 (2015) 116–121

avoidance strategy in which a person perseverates on the causes, consequences, and meaning of the trauma without fully processing the traumatic memory itself (Ehlers and Steil, 1995; Michael et al., 2005). Rumination results in the faulty processing of information and increases PTSD symptom severity (Ehring et al., 2008). Rumination has been found to longitudinally predict PTSD symptoms up to six months following a traumatic event (Ehring et al., 2008). Michael et al.(2007) found that assault survivors with PTSD reported significantly more rumination than assault survivors without PTSD. Furthermore, several types of rumination appear to affect PTSD severity, such as ineffective thinking, perseveration over why something happened and what would have happened had circumstances been different, and continued pressure to ruminate (Michael et al., 2007). 1.2. Rumination and depression Most research investigating rumination has been examined in the context of depression. The Nolen-Hoeksema (1991) Response Styles Theory (RST) of rumination posits that rumination plays a role in onset and maintenance of depression because people who ruminate focus on the causes, meanings, and consequences of their depressive symptoms. Indeed, it appears that rumination exacerbates negative mood in depressed individuals, and increases negative thinking about the future (Lavender and Watkins, 2004). Longitudinal research has shown that rumination predicts the onset of depression in various samples, such as college students (Just and Alloy, 1997), community samples (Huffziger et al., 2009), and adolescents (Wilkinson et al., 2013). A recent meta-analysis by Olatunji et al. (2013) found that participants with depressive disorders reported significantly higher rumination levels than participants with anxiety disorders. In regard to types of rumination, depression has shown a stronger relationship with brooding rumination (comparing one's current situation to an idealized alternative) over reflective rumination (engaging in cognitive problem-solving through internal reflection; see review in Olatunji et al. (2013). 1.3. Rumination in comorbid depression and PTSD To date, there are few studies that have investigated rumination's role in both PTSD and depression. A recent study by Birrer and Michael (2011) found that rumination is heightened in individuals with both PTSD and depression, in contrast to depressed-only individuals; additionally, rumination was a trigger for intrusive images. Rumination also was multidimensional, taking the form of “what-if” types of thoughts, thoughts of depressive content, and thoughts about the original distressing event (Birrer and Michael, 2011).

117

in problem-solving ability and information processing (Tanner et al., 2013). Counterfactual thinking was defined as thoughts about alternative outcomes (Tanner et al., 2013). Tanner et al. (2013) postulated this to be related to feelings of regret, shame, guilt, disappointment, and relief. Repetitive thoughts were defined as repetitive, intrusive, and involuntary thoughts. Anticipatory thoughts were defined as intrusive thoughts and perseveration over future possible events (Tanner et al., 2013). Tanner et al. (2013) purported that this style would be protective against mental disorders because of the possible hopefulness of the cognitions. For example, if someone who is depressed begins ruminating on “in the future I will have a great time with my friends,” s/he may begin to feel better in the moment, and therefore reduce their depressive symptoms. However, this result is in contrast to other findings, which have found that rumination results in more negative future thinking (Lavender and Watkins, 2004); thus, someone who anticipates the future to be bleak may actually have exacerbated depressive symptoms. Second, this study goes beyond past research to investigate subtypes of rumination and their relative impact on the relationship between PTSD and depressive symptoms. Indeed, it is important to recognize that rumination is a complex cognitive construct (Birrer and Michael, 2011) with several facets that can have differential impacts on different disorders. We hypothesized that the four subtypes of rumination from the RTSQ would all significantly moderate the relationship between PTSD and depressive symptoms, meaning that the presence of subtypes of rumination in individuals with PTSD will result in greater depressive symptoms. We believe this for the following reasons: 1) Counterfactual thinking subscale will moderate this relationship, given the evidence for “what-if” types of thinking in both PTSD and depression (Birrer and Michael, 2011), 2) Anticipatory thoughts subscale will moderate this relationship, given evidence that rumination exacerbates negative mood in depressed individuals and leads to increased negative future thinking (Lavender and Watkins, 2004), 3) Repetitive thoughts subscale, given the evidence for intrusive thinking/uncontrollable negative thoughts in both PTSD and depression (Birrer and Michael, 2011), will also moderate this relationship, and 4) Lastly, problemfocused rumination subscale will moderate the relationship, given that models of PTSD (Ehlers and Clark, 2000) posit that rumination results in deficiencies in adaptive problem-solving; and rumination is associated with poor problem-solving in depression (Donaldson and Lam, 2004).

2. Method

1.4. Study aims

2.1. Procedures

Because of the limited research examining rumination's relationships with PTSD and depression, it is important to investigate rumination's impact on these diagnostic constructs. This study is important in several areas. First, it utilized a relatively new measure of rumination, the Ruminative Thought Style Questionnaire (RTSQ; Brinker and Dozois, 2009). This measure is notable because it was initially developed to measure a broader style of ruminative thinking than traditional measures of rumination, and is not biased due to variables such as specific emotional states, the content of items, and information that was time-specific (Brinker and Dozois, 2009). The RTSQ was found to have four subtypes of rumination, including problem-focused thoughts, repetitive thoughts, counterfactual thinking, and anticipatory thoughts (Tanner et al., 2013). Problem-focused thoughts were defined as recurrent thoughts about solving problems, and believed to be associated with deficits

This study was approved by a university Institutional Review Board prior to data collection and was conducted in accordance with the Declaration of Helsinki. Participants seeking mental health services were recruited by trained staff in the waiting room at a community mental health center in the Midwestern U.S. This study utilized a cross-sectional study design. Participants were provided an informed consent statement and then were asked to complete a web survey of questionnaires at their initial intake appointment or second therapy appointment at the clinic. The survey took 30 min to complete and included other measures not assessed in this paper. 2.2. Participants The sample consisted of 45 adults (76% female; 18–71 years old, M¼34.09, SD ¼11.52; 91% Caucasian; 16% Latino) who indicated

118

M.E. Roley et al. / Journal of Affective Disorders 180 (2015) 116–121

having had prior trauma exposure. Participants were selected from a larger sample of 100 individuals consecutively seeking treatment. Participants represented a wide income distribution: 44% made less than $15,000 (n ¼20), 9% made $15,000–$24,999 (n¼ 4), 16% made $25,000–$49,999 (n¼ 7), 13% made $50,000–$79,999 (n ¼6), and 18% made $80,000 or more (n ¼ 8). They represented diverse education levels: 11% had less than a high school diploma (n ¼5), 29% had a high school diploma and no college (n¼ 13), 38% had some college (n ¼17), 11% had a college degree (n ¼5), 9% sought graduate level training (n ¼4), one individual did not report their education level. 2.3. Questionnaires The 20-item RTSQ assessed rumination (e.g., “I tend to replay past events as I would have liked them to happen”) with a 7-point response scale (“1¼ does not describe me at all” to “7¼describes me very well”; Brinker and Dozois, 2009). This study utilized the four subscales of rumination found by the Tanner et al. (2013) factor analysis of the RTSQ. These subscales consisted of the following: problem-focused thoughts (items 1–4; with a 0–28 range of scores), repetitive thoughts (items 5–8; with a 0–28 range of scores), counterfactual thinking (items 9, 11–14; with a 0–35 range of scores), and anticipatory thoughts (items 17 and 20; with a 0–14 range of scores). Only those 15 items were used in the analyses because five items from the original scale did not converge onto their identified factors (Tanner et al., 2013). Cronbach's alphas ranged from .86 to .94, demonstrating excellent internal consistency of the four subscales in the present study. The RTSQ has shown good convergent validity with the Response Style Questionnaire and the Global Rumination Scale (rs¼.396 and.602, respectively), and has shown excellent internal consistency (Brinker and Dozois, 2009). The Patient Health Questionnaire (PHQ-9) is a 9-item selfreport measure of depressive symptoms. Responses range from 0 (“Not at all”) to 3 (“Nearly every day”); higher scores indicate more severity. The PHQ-9 has a high sensitivity and specificity for major depression (Kroenke et al., 2001) as well as excellent internal consistency (α ¼.87; Goodman et al., 1998). In the present study the PHQ-9 had excellent internal consistency (α ¼.93). The Stressful Life Events Screening Questionnaire (Goodman et al., 1998), a self-report measure, assessed for lifetime traumatic event exposure by inquiring about eleven specific traumas and two catch-all items. The response scale has two options (“yes“ or “no“). At the end of the questionnaire, participants were asked which of their endorsed events is causing the most distress. Participants were then instructed to answer the PTSD Checklist for DSM-5 while thinking about the most distressing traumatic event. The SLESQ has shown an 85% specificity rate, as well as excellent concurrent validity and test–retest reliability (Goodman et al., 1998). The PTSD Checklist for DSM-5 (PCL-5) was used to assess PTSD symptoms(Weathers et al., 2013). The PCL-5 is a 20-item measure with a 5-point response scale ranging from 0 ¼“Not at all” to 4 ¼“Extremely.” In this study, internal consistency was excellent (α ¼ .97). The PCL-5 is the latest version of the measure reflecting the DSM-5 diagnostic changes. The initial PCL demonstrated adequate to excellent reliability (α ¼.94; test–retest r ¼.88) and has shown similar psychometric properties across diverse populations including adult civilians seeking outpatient care (Cook et al., 2005). The original PCL's psychometric properties have been reviewed in McDonald and Calhoun (2010) and Wilkins et al. (2011) articles. Preliminary validity work on the PCL-5 has shown that the cut-off score of 38 points reflects probable PTSD (Weathers et al., 2013). In our study, the mean for traumaexposed victims (n ¼45) was 48.31, which suggests that this was a PTSD symptomatic sample.

2.4. Analytic plan First data were screened. Then demographic and trauma exposure data were assessed in order to understand the sample descriptively. The analyses were completed using only the participants who had indicated prior trauma exposure (n¼ 45). All variables were found to be normally distributed. Rumination subscales were found not to be collinear. To assess the hypotheses that the subscales of rumination would moderate the relationship between PTSD and depressive symptoms, four multiple regression analyses were completed using SPSS v.21. Variables were centered, and then interaction terms were calculated prior to entering them into the regression equations. In all of the regressions, PTSD was entered in the first step, followed by a rumination subscale, and in the final step the interaction term was entered. After finding statistically significant moderation effects, the authors probed the moderation analyses to determine at what level the moderation occurred using the MODPROBE SPSS macro (Hayes and Matthes, 2009). 2.5. Exclusions and missing data From the 100 participants who took part in the study, 10 were excluded from analyses in the initial screening phase, 8 of which started the survey but did not complete any measures and two did not fill out the depression measure. Missing data were found to be missing completely at random and were estimated using maximum likelihood (ML) estimation.

3. Results Participants endorsed exposure to a range of traumatic events with an array of traumas that cause the most current distress. Multiple traumas could be endorsed by each participant, but only one trauma could be listed as causing the most current distress. More than half of the sample reported multiple trauma exposures. The most prevalent traumatic events experienced included 80% having a loved one die by accident, suicide, or homicide (n¼ 36), followed by 62% reporting physical abuse (n ¼28), and 56% in a life-threatening accident (n ¼25). Among the most distressing traumatic events reported, 31% of the sample indicated witnessing a loved one die was most distressing (n ¼14), 16% found being in a life-threatening accident was most distressing (n ¼7), and 13% reported being a victim of childhood sexual abuse was most distressing (n ¼6). Results further indicated that 53% (n ¼24) of the sample had a probable PTSD diagnosis based on the DSM-5 PTSD algorithm wherein probable diagnosis persists when one symptom on Cluster B and C and two symptoms on Cluster D are endorsed moderately or higher (Weathers et al., 2013). Thirty-six percent (n ¼16) of the participants had a probable MDE diagnosis based on the DSM-5 criteria that five of the nine items be endorsed at a two or greater, including either the first or second question on the PHQ-9 and four other items (Andrews et al., 2007). The cut-off score used was a total score of 10 or greater. Noteworthy is that 20% (n ¼9) had a comorbid PTSD-MDE diagnosis. Additionally, the overall rumination scale averaged 74.53 (SD ¼ 33.49), with the subscales of problem-focused rumination, counterfactual rumination, repetitive rumination, and anticipatory rumination averaging 16.40 (SD ¼ 8.85), 15.76 (SD ¼7.29), 17.00 (SD ¼7.73), and 6.76 (SD ¼4.07) respectively. 3.1. Multiple regression analyses Results, found in Tables 1 and 2, revealed that only anticipatory rumination and repetitive rumination significantly moderated the

M.E. Roley et al. / Journal of Affective Disorders 180 (2015) 116–121

relationship between PTSD and depressive symptoms. Further clarification of the moderation analyses revealed PTSD significantly predicts depressive symptoms at mean levels of anticipatory rumination (β ¼.030, p ¼.042), as well as at one standard deviation above the mean (β ¼.060, p ¼.008). For repetitive rumination moderation analyses, moderation was only statistically significant at one standard deviation above the mean. That is, at higher levels of repetitive rumination, greater PTSD symptoms predict greater depressive symptoms (β ¼.044, p ¼.016). Problemfocused rumination (β ¼ .112, p¼ .118) and counterfactual rumination (β ¼.160, p ¼.071) did not significantly moderate the relation between PTSD and depression symptoms. There were statistically significant main effects found for both problem-focused rumination (β ¼.736, po.001, R2 ¼ .748) and counterfactual rumination (β ¼.492, p o.001, R2 ¼.526). This means that adults who have greater PTSD symptoms and who engage in repetitive rumination or anticipatory rumination are at greater risk for depressive symptoms, while engagement in other forms of rumination did not increase the risk for depressive symptoms above and beyond having elevated PTSD symptoms. The statistically significant interactions are illustrated in Figs. 1 and 2.

4. Discussion Consistent with our hypotheses, we found that anticipatory rumination and repetitive rumination moderated the relationship between PTSD and depressive symptoms. Inconsistent with our hypotheses, problem-focused rumination and counterfactual rumination did not moderate the relationship between PTSD and depressive symptoms. Anticipatory rumination, defined as intrusive thoughts and perseveration over future possible events, moderated the relationship between PTSD and depressive symptoms at the mean and one standard deviation above the mean level of anticipatory rumination. This means that when anticipatory rumination is at an average level or higher and someone has PTSD symptoms, they are likely to also have greater depressive symptoms. We argue that anticipatory rumination among Table 1 Regression analysis of repetitive rumination moderating the relationship between PTSD and depression. Predictors Step 1 PTSD Step 2 PTSD Repetitive rumination Step 3 PTSD  repetitive rumination n

B

SE B

β

t

p

ΔR2

.030

.016

.189

1.905

.06

.036

.015 .593

.012 .072

.090 .622

1.577 8.161

.118 .000

.394nn

.003

.001

.171

2.239

.027

.028n

po .05. p o .01.

nn

Table 2 Regression analysis of anticipatory rumination moderating the relationship between PTSD and depression. Predictors Step 1 PTSD Step 2 PTSD Anticipatory rumination Step 3 PTSD  anticipatory rumination n

p o .05. p o .01.

nn

B

SE B

β

t

p

ΔR2

.030

.016

.189

1.905

.060

.036

.024 .834

.014 .167

.147 .441

2.060 4.849

.105 .000

.193nn

.008

.004

.193

2.154

.034

.036n

119

individuals with PTSD who have intrusive memories about their traumatic events, which are painful and shocking to experience and can happen at any time, may exacerbate co-occurring depressive symptoms. That is, if individuals ruminate on the trauma or on the possibility of more traumatic experiences happening in the future, they may feel more depressed. Though the causal nature of the relationship between PTSD symptoms, rumination, and depressive symptoms could not be assessed in the current study, thus we are only able to conclude that when individuals have intrusive PTSD symptoms, and anticipatory rumination at moderate and high levels, co-occuring depressive symptoms are elevated. This is consistent with Lavender and Watkins' (2004) findings. Further, research has found that individuals with PTSD have altered worldviews wherein they view themselves, others, and the world negatively (Fischer and Ayoub, 1994; Iverson et al., 2011; JanoffBulman and Frieze, 1983). Likewise, people with depression tend to have more negative views of themselves and their future (Iverson et al., 2011). Thus, if someone ruminates about the possibility of more negative future events, his/her depressive symptoms would be greater than if they did not have PTSD. Tanner et al. (2013) hypothesized that anticipatory rumination would be protective against depressive symptoms. Our results show the contrary when PTSD and depression symptoms co-occur. In a longitudinal study examining rumination's predictive ability for depression in a match-paired sample of formerly hospitalized and community adults, Huffziger et al. (2009) found that across both samples, individuals who ruminate about their depressive symptoms are at higher risk for future depressive episodes. This lends support to the current authors' view that anticipatory rumination may be tapping into rumination about depressive symptoms, which appears to put people at higher risk for problems. Repetitive rumination, consistent with more traditional conceptualizations of rumination (Nolen-Hoeksema et al., 1993), is repetitive and intrusive thoughts that occur involuntarily (Brinker and Dozois, 2009). Our analyses found that repetitive rumination only moderates the relationship between PTSD and depression when repetitive rumination is high. As discussed, the intrusive nature of traumatic memories lends itself to more depressogenic thoughts. It would make sense that someone with higher levels of repetitive rumination would have worse PTSD symptoms and depressive symptoms given the overlap in dysphoric symptoms between the two disorders (Biehn et al., 2013; Contractor et al., 2014; Elklit et al., 2010). Additionally, cognitive symptoms associated with PTSD may play an important role with repetitive rumination. Cognitive symptoms include when a person's view of the world, themselves, or others is altered negatively; in accord with the Theory of Shattered Assumptions (Janoff-Bulman and Frieze, 1983), such that after trauma exposure they view the world as scarier and more dangerous. If individuals with PTSD have the cognitive symptoms, repetitive rumination may serve to exacerbate those negatively altered cognitions. Regarding our findings that were contrary to our hypotheses, it may be that problem-focused rumination is impairing at all levels of PTSD and depressive symptoms. Thus, it may be better understood as having a direct effect on problem-solving ability and information processing, as opposed to an indirect effect. Indeed, studies have found evidence for the direct impact of rumination on problem-solving ability (Lyubomirsky and Nolen-Hoeksema, 1995; Nolen-Hoeksema et al., 2008). We too found problem-focused rumination to have a direct effect on both PTSD and depression symptoms. Counterfactual thinking, defined as thoughts about alternative outcomes, was projected to relate to feelings of regret, guilt, and relief (Tanner et al., 2013) by the authors of the RTSQ. It is possible that individuals who are depressed have a difficult time engaging in thinking about alternative outcomes. In this vein, research has found that depressed individuals tend to get stuck in a “tunnel vision.” Thus, while research has found individuals with PTSD to have a high level of guilt/regret (Okulate and Jones, 2006), it may be that those with cooccurring depression do not engage in counterfactual thinking.

120

M.E. Roley et al. / Journal of Affective Disorders 180 (2015) 116–121

Fig. 1. Repetitive rumination moderates relationship between PTSD and depressive symptoms.

Fig. 2. Anticipatory rumination moderates relationship between PTSD and depressive symptoms.

4.1. Implications

4.2. Limitations and future research

There are important clinical implications stemming from this study. First, our findings highlight the need for assessment of rumination, especially anticipatory and repetitive rumination in patients with comorbid PTSD and depressive symptoms. Second, it will be important for interventions to target anticipatory rumination and repetitive rumination in the treatment of co-occurring PTSD and depressive symptoms. Specifically, cognitive-behavioral interventions, which successfully reduce depressive symptoms (Feng et al., 2012; Hans and Hiller, 2013; Vittengl et al., 2007), however are 81% effective at reducing relapse in the first year and only 46% effective in reducing relapse in the second year posttreatment (Vittengl et al., 2007). For individuals who ruminate, cognitive-behavioral interventions can focus on reducing rumination in order to reduce the barriers to effective cognitive restructuring. Thus, rumination, which maintains/exacerbates depressive thought cycles, can be targeted by disputing beliefs on the utility of rumination. Once rumination is reduced, thought restructuring can then be more effectively targeted. Finally, results highlight the importance of assessing for subscales of rumination rather than the overall rumination construct given differences in findings associated with different rumination subscales.

Our study had limitations. Although it was a strength of our study to collect data from a community-based clinical population, our results should be replicated with other trauma populations (for example, military war veterans) due to the possibility that the number and complexity of traumatic events may impact PTSD symptoms, rumination, and co-occurring depressive symptoms. Another limitation is that we used the PCL-5, which has preliminary validity with only forty-five participants, thus the measure's generalizability may be limited, however this measure is the only known validated measure of PTSD for the DSM-5 (Weathers et al., 2013). The PCL-5 has shown excellent psychometric properties (Weathers et al., 2013), and more research is needed to confirm these validity findings. Further, we were unable to assess the content of participants' rumination. Thus, we are unable to confirm why rumination styles moderated the relationship between PTSD and depression or why counterfactual thinking and problem-focused thinking did not moderate the relationship between the two constructs. Future research should assess the content of participants' rumination as well as the style of rumination that participants' engage in. Lastly, we used self-report measures that may lead to response biases and social desirability effects. Future research can benefit from multimethod assessments.

M.E. Roley et al. / Journal of Affective Disorders 180 (2015) 116–121

Role of funding source The research presented in this manuscript was unfunded. The University of Toledo Psychology Department paid the salaries of Meredith Claycomb, Ateka Contractor, and Jon Elhai. The University of Toledo College of Graduate Studies paid Michelle Roley in the form of a University Fellowship. Ateka Contractor is currently being paid by the Department of Psychiatry & Human Behavior, Warren Alpert Medical School of Brown University. Cherie Armour's salary is paid for by the Psychiatric Research Institute, Ulster University. Paula Dranger's salary is paid for by Choices! Counseling Services.

Conflict of interest Author declarations We wish to draw the Editor's attention to the following facts which may be considered as potential conflicts of interest and to financial contributions to this work. One coauthor, Jon Elhai, received financial support in the form of personal fees from Elsevier Foundation, WILEY, Other (occasional fees as an expert witness in court), and the Menninger Clinic. He also received financial support in the form of a Grant from DoD (W81XHW-07-1-0409) and NIH (1R21MH098198-01A1) accounting for 5% of his salary. No other co-authors have potential conflicts of interest to disclose and there has been no significant financial support for this work that could have influenced its outcome. We confirm that the manuscript has been read and approved by all named authors and that there are no other persons who satisfied the criteria for authorship but are not listed. We further confirm that the order of authors listed in the manuscript has been approved by all coauthors. We confirm that we have given due consideration to the protection of intellectual property associated with this work and that there are no impediments to publication, including the timing of publication, with respect to intellectual property. In so doing we confirm that we have followed the regulations of our institutions concerning intellectual property. We further confirm that aspects of the work covered in this manuscript that has involved human patients has been conducted with the ethical approval of all relevant bodies as acknowledged within the manuscript.

Acknowledgments The authors of this manuscript would like to thank the participants who took part in this study. Without their willingness to participate as well as thoughtfulness in responding to questionnaires this project would not have been possible. Further, the authors would like to acknowledge the hard work and dedication that the staff at Choices! Counseling Services has put forth. Without their team effort, this project would not have come to fruition.

References Andrews, G., Brugha, T., Thase, M.E., Duffy, F.F., Rucci, P., Slade, T.M., 2007. Dimensionality and the category of major depressive episode. Int. J. Methods Psychiatr. Res. 16 (S1), S41–S51. Biehn, T.L., Contractor, A., Elhai, J.D., Tamburrino, M., Fine, T.H., Prescott, M.R., et al., 2013. Relations between the underlying dimensions of PTSD and major depression using an epidemiological survey of deployed Ohio National Guard soldiers. J. Affect. Disord. 144 (1–2), 106–111. Birrer, E., Michael, T., 2011. Rumination in PTSD as well as in traumatized and nontraumatized depressed patients: a cross-sectional clinical study. Behav. Cogn. Psychother. 39 (4), 381–397. Brewin, C.R, Holmes, E.A., 2003. Psychological theories of posttraumatic stress disorder. Clin. Psychol. Rev. 23 (3), 339–376. Brinker, J.K., Dozois, D.J., 2009. Ruminative thought style and depressed mood. J. Clin. Psychol. 65 (1), 1–19. Contractor, A.A., Durham, T.A., Brennan, J.A., Armour, C., Wutrick, H.R., Frueh, B.C., et al., 2014. DSM-5 PTSD's symptom dimension and relations with major depression's symptom dimensions in a primary care sample. Psychiatry Res. 215, 146–153. Cook, J.M., Elhai, J.D., Arean, P.A., 2005. Psychometric properties of the PTSD Checklist with older primary care patients. J. Trauma. Stress 18 (4), 371–376. Donaldson, C., Lam, D., 2004. Rumination, mood, and social problem-solving in major depression. Psychol. Med. 34 (7), 1309–1318. Ehlers, A., Clark, D.M., 2000. A cognitive model of posttraumatic stress disorder. Behav. Res. Ther. 38 (4), 319–345. Ehlers, A., Steil, R., 1995. Maintenance of intrusive memories in posttraumatic stress disorder: a cognitive approach. Behav. Cogn. Psychother. 23 (3), 217–249. Ehring, T., Frank, S., Ehlers, A., 2008. The role of rumination and reduced concreteness in the maintenance of posttraumatic stress disorder and depression following trauma. Cognit. Ther. Res. 32 (4), 488–506. Elhai, J.D., Grubaugh, A.L., Kashdan, T.B., Frueh, B.C., 2008. Empirical examination of a proposed refinement to DSM-IV posttraumatic stress disorder symptom

121

criteria using the National Comorbidity Survey Replication data. J. Clin. Psychiatry 69 (4), 597–602. Elklit, A., Armour, C., Shevlin, M., 2010. Testing alternative factor models of PTSD and the robustness of the dysphoria factor. J. Anxiety Disord. 24, 147–154. Elwood, L.S., Hahn, K.S., Olatunji, B.O., Williams, N.L., 2009. Cognitive vulnerabilities to the development of PTSD: a review of four vulnerabilities and the proposal of an integrative vulnerability model. Clin. Psychol. Rev. 29 (1), 87–100. Feng, C.Y., Chu, H., Chen, C.H., Chang, Y.S., Chen, T.H., Chou, Y.H., et al., 2012. The effect of cognitive behavioral group therapy for depression: a meta‐analysis 2000–2010. Worldviews Evid. Based Nurs. 9 (1), 2–17. Fischer, K.W., Ayoub, C., 1994. Affective splitting and dissociation in normal and maltreated children: developmental pathways for self in relationships. In: Cicchetti D., Toth S., (Eds.), Proceedings of the Rochester Symposium on Developmental Psychopathology, Vol. 5: The self and its disorders, Vol. 5. Rochester, N.Y., University of Rochester Press, pp. 149–222. Goodman, L.A., Corcoran, C., Turner, K., Yuan, N., Green, B.L., 1998. Assessing traumatic event exposure: general issues and preliminary findings for the Stressful Life Events Screening Questionnaire. J. Trauma. Stress 11 (3), 521–542. Grubaugh, A.L., Long, M.E., Elhai, J.D., Frueh, B.C., Magruder, K.M., 2010. An examination of the construct validity of posttraumatic stress disorder with veterans using a revised criterion set. Behav. Res. Ther. 48 (9), 909–914. Hans, E., Hiller, W., 2013. Effectiveness of and dropout from outpatient cognitive behavioral therapy for adult unipolar depression: a meta-analysis of nonrandomized effectiveness studies. J. Consult. Clin. Psychol. 81 (1), 75–88. Hayes, A.F., Matthes, J., 2009. Computational procedures for probing interactions in OLS and logistic regression: SPSS and SAS implementations. Behav. Res. Methods 41, 924–936. Huffziger, S., Reinhard, I., Kuehner, C., 2009. A longitudinal study of rumination and distraction in formerly depressed inpatients and community controls. J. Abnorm. Psychol. 118 (4), 746–756. Iverson, K.M., Gradus, J.L., Resick, P.A., Suvak, M.K., Smith, K.F., Monson, C.M., 2011. Cognitive-behavioral therapy for PTSD and depression symptoms reduces risk for future intimate partner violence among interpersonal trauma survivors. J. Consult. Clin. Psychol. 79 (2), 193–202. Janoff-Bulman, R., Frieze, I.H., 1983. A theoretical perspective for understanding reactions to victimization. J. Soc. Issues 39 (2), 1–17. Just, N., Alloy, L.B., 1997. The response styles theory of depression: tests and an extension of the theory. J. Abnorm. Psychol. 106 (2), 221–229. Kroenke, K., Spitzer, R.L., Williams, J.B., 2001. The PHQ-9: validity of a brief depression severity measure. J. Gen. Intern. Med. 16 (9), 606–613. Lavender, A., Watkins, E., 2004. Rumination and future thinking in depression. Br. J. Clin. Psychol. 43 (Pt 2), 129–142. Lyubomirsky, S., Nolen-Hoeksema, S., 1995. Effects of self-focused rumination on negative thinking and interpersonal problem solving. J. Personal. Soc. Psychol. 69 (1), 176–190. McDonald, S.D., Calhoun, P.S., 2010. The diagnostic accuracy of the PTSD checklist: a critical review. Clin. Psychol. Rev. 30 (8), 976–987. Michael, T., Ehlers, A., Halligan, S.L., Clark, D.M., 2005. Unwanted memories of assault: what intrusion characteristics are associated with PTSD? Behav. Res. Ther. 43 (5), 613–628. Michael, T., Halligan, S.L., Clark, D.M., Ehlers, A., 2007. Rumination in posttraumatic stress disorder. Depress. Anxiety 24 (5), 307–317. Nolen-Hoeksema, S., 1991. Responses to depression and their effects on the duration of depressive episodes. J. Abnorm. Psychol. 100 (4), 569–582. Nolen-Hoeksema, S., Morrow, J., Fredrickson, B.L., 1993. Response styles and the duration of episodes of depressed mood. J. Abnorm. Psychol. 102 (1), 20–28. Nolen-Hoeksema, S., Wisco, B.E., Lyubomirsky, S., 2008. Rethinking Rumination. Perspect. Psychol. Sci. 3 (5), 400–424. Okulate, G.T., Jones, O.B., 2006. Post-traumatic stress disorder, survivor guilt and substance use–a study of hospitalised Nigerian army veterans. S. Afr. Med. J. 96 (2), 144–146. Olatunji, B.O., Naragon-Gainey, K., Wolitzky-Taylor, K.B., 2013. Specificity of rumination in anxiety and depression: a multimodal meta-analysis. Clin. Psychol.: Sci. Pract. 20 (3), 225–257. Rippere, V., 1977. Some cognitive dimensions of antidepressive behaviour. Behav. Res. Ther. 15 (1), 57–63. Tanner, A., Voon, D., Hasking, P., Martin, G., 2013. Underlying structure of ruminative thinking: factor analysis of the ruminative thought style questionnaire. Cognit. Ther. Res. 37 (3), 633–646. Vittengl, J.R., Clark, L.A., Dunn, T.W., Jarrett, R.B., 2007. Reducing relapse and recurrence in unipolar depression: a comparative meta-analysis of cognitivebehavioral therapy's effects. J. Consult. Clin. Psychol. 75 (3), 475–488. Weathers, F.W., Litz, B.T., Keane, T.M., Palmieri, P.A., Marx, B.P., Schnurr, P.P., 2013. The PTSD Checklist for DSM-5 (PCL-5). National Center for PTSD. Wilkins, K.C., Lang, A.J., Norman, S.B., 2011. Synthesis of the psychometric properties of the PTSD checklist (PCL) military, civilian, and specific versions. Depress. Anxiety 28 (7), 596–606. Wilkinson, P.O., Croudace, T.J., Goodyer, I.M., 2013. Rumination, anxiety, depressive symptoms and subsequent depression in adolescents at risk for psychopathology: a longitudinal cohort study. BMC Psychiatry 13, 250.