Journal of Anxiety Disorders 28 (2014) 67–74

Contents lists available at ScienceDirect

Journal of Anxiety Disorders

Posttraumatic stress disorder and alcohol dependence: Individual and combined associations with social network problems夽 Courtney E. Dutton a,∗ , Thomas Adams a,b,c , Sarah Bujarski a , Christal L. Badour a,b,c , Matthew T. Feldner a,d,∗∗ a

University of Arkansas, United States Medical University of South Carolina, United States c Ralph H. Johnson Veterans Affairs Medical Center, United States d Laureate Institute for Brain Research, United States b

a r t i c l e

i n f o

Article history: Received 6 June 2013 Received in revised form 15 November 2013 Accepted 30 November 2013 Keywords: PTSD Alcohol dependence Social support Social conflict

a b s t r a c t People with either posttraumatic stress disorder (PTSD) or alcohol dependence (AD) are apt to report problems in their social networks, including low perceived support and elevated conflict. However, little research has examined social networks among people with comorbid PTSD/AD despite evidence suggesting these two conditions commonly co-occur and are linked to particularly severe problems. To test the hypothesis that people with comorbid PTSD/AD experience particularly elevated social network problems, individuals with lifetime diagnoses of PTSD, AD, comorbid PTSD/AD, or no lifetime history of Axis I psychopathology in the National Comorbidity Survey-Replication were compared on four dimensions of social networks: (1) Closeness, (2) Conflict, (3) Family Support, and (4) Apprehension. Persons with PTSD, AD, or comorbid PTSD/AD endorsed more problems with the Conflict, Family Support, and Apprehension factors compared to people with no history of Axis I psychopathology. Moreover, individuals with comorbid PTSD/AD endorsed greater Apprehension and significantly less Family Support compared to the other three groups. Results suggest people with comorbid PTSD/AD experience increased problems with their family as well as greater concerns about enlisting social support than even people with PTSD or AD alone. Treatments for people suffering from comorbid PTSD/AD should consider assessing for and possibly targeting family support and apprehension about being close to others. © 2013 Elsevier Ltd. All rights reserved.

1. Introduction Posttraumatic stress disorder (PTSD) and alcohol dependence (AD) are chronic, often disabling conditions (Hasin, Stinson, Ogburn, & Grant, 2007; Kessler, 2000; Kessler, Sonnega, Bromet, Hughes, & Nelson, 1995; Samokhvalov, Popova, Room, Ramos, & Rehm, 2010). Moreover, PTSD and AD are commonly comorbid (Back, Jackson, Sonne, & Brady, 2005; Brown, Recupero, & Stout, 1995; Kessler et al., 1997; Pietrzak, Goldstein, Southwick, & Grant,

夽 The views expressed in this manuscript are those of the authors and do not necessarily represent those of the Department of Veterans Affairs or the US government. ∗ Corresponding author at: Department of Psychological Science, University of Arkansas, 216 Memorial Hall, Fayetteville, AR 72701, United States. Tel.: +1 479 575 5811; fax: +1 479 575 3219. ∗∗ Corresponding author at: Department of Psychological Science, University of Arkansas, 216 Memorial Hall, Fayetteville, AR 72701, United States. Tel.: +1 479 575 5817; fax: +1 479 575 3219. E-mail addresses: [email protected] (C.E. Dutton), [email protected] (M.T. Feldner). 0887-6185/$ – see front matter © 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.janxdis.2013.11.010

2011; Stewart, 1996). Research suggests that compared to the presence of unimorbid PTSD or AD, comorbid PTSD/AD is associated with greater problem severity across multiple indices. For example, the co-occurrence of these two disorders is associated with a more severe clinical presentation (Mills, Teeson, Ross, & Peters, 2006; Ouimette, Goodwin, & Brown, 2006), including increased cooccurrence of additional anxiety disorders and depression (Bonin, Norton, Asmundson, Dicurzio, & Pidlubney, 2000; Drapkin et al., 2012; Rash, Coffey, Baschnagel, Drobes, & Saladin, 2008; Read, Brown, & Kahler, 2004), a longer history of problematic substance use, a greater likelihood of suicide attempts (Bonin et al., 2000), and worse treatment outcomes (e.g., Brown & Wolfe, 1994; Najavits, Weiss, & Shaw, 1999; Ouimette, Brown, & Najavits, 1998; Ouimette, Finney, & Moos, 1999). As such, it is important to identify factors that differ between people with comorbid PTSD/AD and those suffering from only one of these problems in order to advance our currently limited understanding of what may account for the particularly severe problems introduced by this comorbidity. People with PTSD experience elevated problems with social networks. There is a strong negative association between social support and symptoms of posttraumatic stress subsequent to

68

C.E. Dutton et al. / Journal of Anxiety Disorders 28 (2014) 67–74

traumatic event exposure (Brewin, Andrews, & Valentine, 2000; Danner & Radnitz, 2000; Eriksson, Vande Kemp, Gorsuch, Hoke, & Foy, 2001; Kaniasty & Norris, 2008; Tucker, Pfefferbaum, Nixon, & Dickson, 2000). Research has consistently demonstrated a bidirectional association between PTSD severity and social support (Kaniasty & Norris, 2008; Turner, 1981). For example, one study found that pre-trauma social support predicted PTSD symptoms 6–12-months following traumatic event exposure, while both pathways from social support to PTSD, as well as PTSD to social support were significant 12–18 months following the traumatic event (Kaniasty & Norris, 2008). Symptoms of PTSD, such as loss of interest in activities, feelings of estrangement of others, and increased anger are possible mechanisms underlying decreases in social support following a traumatic event (Kaniasty & Norris, 2008). People with PTSD also endorse elevated levels of social conflict (Galovski & Lyons, 2004; Monson, Taft, & Fredman, 2009; Monson et al., 2012). Indeed, PTSD has been associated with relatively elevated levels of both physical and psychological aggression in significant interpersonal relationships (Taft, Watkins, Stafford, Street, & Monson, 2011). For example, in a sample of treatment seeking male veterans with a partner, approximately 33% endorsed perpetrating physical aggression, and 91% endorsed perpetrating psychological aggression toward their partner (Taft, Weatherill et al., 2009). Similarly, PTSD symptoms were positively correlated with both physical and psychological victimization and perpetration in a civilian sample of flood survivors (Taft, Monson et al., 2009). Deficits in social support have also been linked to alcohol use. Research suggests people with supportive friends and families report more success in reducing alcohol use (Beattie & Longabaugh, 1997; Gordon & Zrull, 1991; Tucker, Vuchicich, & Pukish, 1995). Social support for alcohol-related treatment and abstinence, more specifically, is positively associated with percentage of days abstinent following treatment, and negatively correlated with the proportion of days of heavy drinking following treatment (Beattie & Longabaugh, 1999). Research on alcohol use disorders and relationship functioning has suggested an association between alcohol use and relationship difficulties, including lower marital satisfaction (Marshal, 2003) and elevated marital aggression (Leonard & Blane, 1992; Murphy & O’Farrell, 1994). Similarly, effects of drinking in men with an alcohol use disorder have been linked to elevated social conflict (Kachadourian, Taft, O’Farrell, Doron-LaMarca, & Murphy, 2012; O’Farrell & Murphy, 1995). In fact, social conflict has been associated with the maintenance of alcohol use via its correlation with relapse during alcohol use quit attempts (Marlatt & Gordon, 1980). Despite a corpus of data suggesting problems with social networks are linked to both PTSD and AD, relatively little research has examined social network problems among people with comorbid PTSD/AD. Preliminary evidence suggests that people with PTSD/AD may experience particularly severe problems with social networks. People with comorbid PTSD/AD are less likely to be married (Drapkin et al., 2012) and more likely to report interpersonal problems (Najavits et al., 1998). Given these data, the current study tested if the combination of PTSD and AD is related to particularly elevated social network problems, even relative to each of these conditions alone. First, we predicted that people with either lifetime PTSD, AD, or comorbid PTSD/AD would report lower levels of perceived social closeness to others, decreased ability to rely on or open up to relatives, greater social conflict, and more apprehension about utilizing social support when compared to those without a lifetime history of Axis I psychopathology. Second, it was predicted that individuals with lifetime comorbid PTSD/AD would report lower perceived social support in the form of lower perceived closeness in general, as well as reduced perceived ability to rely on or open up to relatives, greater perceived social conflict, and more apprehension about social support when

compared to people who meet lifetime criteria for only one of these disorders. After first examining the predicted associations between PTSD and AD with these social network factors, these associations were examined after statistically covarying for variance accounted for by other comorbid conditions. Given both PTSD and AD are commonly comorbid with multiple types of psychopathology that also may impact social networks (e.g., major depressive disorder, panic disorder; Back, Jackson, Sonne, & Brady, 2005; Bonin, Norton, Asmundson, Dicurzio, & Pidlubney, 2000; Drapkin et al., 2012; Jacobsen, Southwick, & Kosten, 2001; Rash, Coffey, Baschnagel, Drobes, & Saladin, 2008; Stewart, 1996), it is important to try and gauge the degree to which comorbid PTSD/AD is uniquely related to social network problems above and beyond other comorbid conditions. While analysis of covariance does not equate groups given important group differences (Miller & Chapman, 2001), it can be used to tentatively gauge the uniqueness of an association given inferences are situated within the limitations of the approach (Zinbarg, Suzuki, Uliaszek, & Lewis, 2010).

2. Method 2.1. Participants The current study examined data from the National Comorbidity Survey-Replication (NCS-R; Kessler & Merikangas, 2004). This nationwide epidemiological study included a nationallyrepresentative sample of English-speaking adults (≥17 years old) from 48 states in the United States. To identify potential participants, a stratified, multistage probability sample was utilized. Participation in Part I was adjusted for the differential probability of selection between the sample and the United States population with regard to sociodemographic and geographic variables. The NCS-R procedure employed a structured research interview (Kessler et al., 2004) to evaluate psychiatric disorders as categorized in the Diagnostic and Statistical Manual–Fourth Edition (DSM-IV; American Psychiatric Association [APA], 1994). Interviews were administered by researchers in participants’ homes between February 2001 and April 2003. There was a 73% response rate for the interview. See Kessler et al. (2004) for a detailed description of sampling procedures and methods. The structured interview consisted of two parts. Part I was comprised of a core diagnostic interview that was administered to all participants (n = 9282). More specifically, the World Health Organization’s Composite International Diagnostic Interview (CIDI; Kessler et al., 2004) was employed to index psychopathology diagnoses. Part I was completed in approximately one hour. Part II was then administered to a subsample of participants (n = 5692). Part II more explicitly focused on specific types of psychopathology (e.g., PTSD), as well as correlates and risk factors for psychopathology (e.g., social support measures). Participation in Part II of the study was determined based on the presence of at least one of the following criteria: (1) lifetime diagnosis of any disorder in Part I, (2) meeting subthreshold criteria for a DSM-IV-defined Axis I disorder, (3) ever having sought treatment for a psychological disorder, (4) use of psychotropic medications in the past 12-months, or (5) lifetime history of attempted suicide or a plan to commit suicide. A total of 59% of individuals meeting at least one of these criteria participated in Part II. Finally, in addition to individuals selected for Part II based on these criteria, a probability subsample of other respondents was also included in Part II; thus, an additional 25% of the entire sample was also selected to participate in Part II (Kessler et al., 2004). All portions of the interview were administered by professionally trained interviewers who were closely supervised by the

C.E. Dutton et al. / Journal of Anxiety Disorders 28 (2014) 67–74

69

Table 1 Sample descriptive information. Diagnostic group No Axis I Diagnosis (n = 2180.5) M (SE) or %

PTSD (n = 303.9) M (SE) or %

AD (n = 224.8) M (SE) or %

PTSD/AD (n = 56.4) M (SE) or %

Age** Sex (female)**

48.13 (.765) 51.4%

41.67 (.744)a 79.6%a,b

40.38 (.898)a 25.2%a,b

40.59 (1.011)a 56.7%

Race/ethnicity Hispanic Black White Other

12.4% 13.6% 70.8% 3.2%

8.1% 14.6% 72.8% 4.5%

11.7% 7.3% 75.5% 5.5%

18.4% 6.3% 70.1% 5.1%

Education (years) Annual income*

13.11 $60,922.16

13.24 $53,482.53a

12.88 $53,408.70a

12.80 $53,513.93

Marital status Married/cohabitating Separated/widowed/divorced Never married

60.9% 18.8% 20.3%

50.9% 27.0% 22.1%

51.3% 20.8% 27.9%

49.3% 29.2% 21.5%

Note: *p ≤ .05, **p ≤ .001 indicates group differences across diagnostic groups. a Indicates diagnostic group is significantly different than No Axis I Diagnosis group. b Indicates diagnostic group is significantly different than PTSD/AD group.

Institute for Social Research at the University of Michigan, Ann Arbor. Participants were compensated $50 for their time. The current study examined a subset of participants in Part II of the NCS-R who met one of the following inclusion criteria: (1) no lifetime diagnosis of an Axis I disorder, (2) met criteria for a lifetime diagnosis of PTSD, (3) met criteria for a lifetime diagnosis of AD, or (4) met criteria for lifetime, comorbid PTSD/AD (i.e., met criteria for both PTSD and AD at some point in their lifetime). Individuals were also required to have full data for a measure of social support administered during Part II of the study. Table 1 includes descriptive information for the sample. 2.2. Measures 2.2.1. Composite International Diagnostic Interview (CIDI; Kessler & Üstün, 2004) The CIDI was used to index DSM-IV-defined anxiety disorders, impulse control disorders, mood disorders, and substance use disorders. The National Comorbidity Study-Replication divided the administration of the CIDI into two parts to address the length of the interview. Of note, alcohol dependence was assessed during Part I of the interview while assessment of PTSD was included in Part II because it required extensive introductory questions. 2.2.2. Social Network Problems Section 34 of the NCS-R was comprised of 15 items created to measure problems related to social networks (National Comorbidity Survey, n.d.). The current study omitted item 12 (“When you have a problem or worry, how often do you let your [husband/wife/partner] know about it?”) because only respondents with a husband/wife/partner responded to this item.1 We were interested in examining social network problems among people both in and outside of such relationships. Items were rated on a Likert-type scale; however, the range and anchor descriptors varied by item (e.g., 1 = A Lot or Often, 2 = Some or Sometimes, 3 = A Little or Rarely, 4 = Not At All or Never). See Table 2 for an enumeration of all items. For the current study, items were recoded such that higher scores are indicative of poorer social functioning (e.g., greater social conflict, less perceived support).

1 Item 11 in the Social Networks measure was a reviewer checkpoint and therefore neither relevant to, nor included in, the current study.

2.3. Data analytic approach A principle components analysis (PCA) was performed on the Social Network items. Analyses of variance (ANOVAs) were then conducted with the SPSS 19 Complex Samples module to compare individuals with lifetime PTSD, AD, PTSD/AD, or no lifetime Axis I diagnosis in terms of the social network problem factors identified via the PCA. Planned comparisons were then conducted to evaluate significant differences between diagnostic groups. Finally, these primary analyses were then repeated after controlling for demographic factors that significantly differed across groups as well as other types of psychopathology significantly associated with the social network factor being examined in a given analysis.

3. Results 3.1. Examination of social network items First, the 14 items from the social network measure were simultaneously submitted to a PCA. The first four eigenvalues from the PCA were 3.18 (Closeness), 2.31 (Conflict), 1.29 (Family Support), and 1.10 (Apprehension), which accounted for 56.29% of total variance. Parallel analysis supported a 4-factor solution, as eigenvalues of the four PCA extracted factors were all larger than the upper limit of the 95% confidence interval [CI] for their parallel Monte Carlo generated eigenvalues (M = 1.13, 1.11, 1.08, 1.06; respectively). All items adequately loaded onto their respective factor and there were no complex loadings. Table 2 presents the pattern matrix (i.e., loadings) and communalities for the four-component solution. Given that factors were comprised of varying numbers of items, factor scores were calculated by computing the average within each factor. Component I was comprised of five items (M = 2.17, SD = .81). This component was labeled “Closeness” as each item asks about the degree to which respondents feel close to, or are able to open up to, others. Component II was comprised of four items (M = 3.20, SD = 1.33). This component was labeled “Conflict” as each item asks about frequency of arguments and perceived demands placed on respondents. Component III was comprised of three items (M = 4.33, SD = 2.04). This component was labeled “Family Support” as each item asks about degree of support respondents obtain from relatives. Component IV was comprised of two items (M = 1.47, SD = .65). This component was labeled “Apprehension”

70

C.E. Dutton et al. / Journal of Anxiety Disorders 28 (2014) 67–74

Table 2 Principal components analysis of the National Comorbidity Survey-Replication social networks interview items: loadings, communalities (h2 ), means (M), and standard deviations (SD) for the 4-factor solution. Social network items 8. How much can you open up to your friends if you need to talk about your worries? 7. How much can you rely on your friends for help if you have a serious problem? 6. How often do you talk on the phone or get together with friends? 13. When you have a problem or worry, how often do you let someone (else) know about it? 14. I find it relatively easy to get close to other people. 5. How often do your relatives argue with you? 9. How often do your friends make too many demands of you? 4. How often do your relatives make too many demands on you? 10. How often do your friends argue with you? 2. How much can you rely on relatives who do not live with you for help if you have a serious problem? 3. How much can you open up to relatives who do not live with you if you need to talk about your worries? 1. How often do you talk on the phone or get together with relatives who do not live with you? 16. I find that others are reluctant to get as close to me as I would like. 15. I am somewhat uncomfortable being close to others.

Closeness

Conflict

Family Support

Apprehension

h2

M

SD

.85

.03

.01

.12

.45

2.66

1.32

.83

−.06

−.01

−.05

.69

1.61

.95

.73

.09

−.07

.01

.69

1.86

1.02

.45

−.04

.17

.12

.56

3.16

.92

.38 −.07 .09

.02 .73 .73

.15 .08 −.09

−.29 .07 −.10

.56 .52 .69

3.30 2.57 1.86

.83 1.31 1.02

−.15

.72

.18

.08

.72

1.91

1.00

.15 .00

.69 −.00

−.14 .82

.00 −.06

.54 .52

3.39 3.49

.75 .68

.12

−.05

.76

−.09

.27

3.09

1.09

−.01

.05

.68

.04

.33

2.07

1.03

.14

−.02

−.00

.84

.66

3.31

.93

−.10

.09

−.05

.76

.68

3.77

.60

Note: Bold values are indicative of social network variables that loaded on to that specific social network factor.

as both items ask about the degree of distress/discomfort/anxiety about getting close to others. In terms of internal consistency of the four social network factors, Cronbach’s alpha was adequate for the Closeness (˛ = .71), Conflict (˛ = .69), and Family Support (˛ = .65) factors, but low for the Apprehension factor (˛ = .50). The Apprehension factor evidenced the poorest internal consistency. This was possibly due to the fact that Cronbach’s alpha values are inversely related to the number of items used to estimate the statistic (e.g., alpha is higher when k is large; Raykov, 1997). As such, Cronbach’s alpha has limited utility as a test statistic for the Apprehension scale. 3.2. Descriptive statistics Complex samples general linear modeling revealed significant group differences in annual income [F(3, 40) = 3.255, p = .031] and age [F(3, 40) = 26.945, p < .001], and logistic regression conducted in the complex samples module indicated differences in proportion of women across groups (Wald F = 59.262, p < .001). See Table 1 for descriptive information regarding these factors. There were no significant group differences in the remaining demographic variables. As such, age, annual income, and sex were entered as covariates in secondary analyses of covariance (ANCOVAs) examining social network factors as a function of PTSD and alcohol dependence to estimate potentially unique associations among these variables. Complex samples general modeling also was utilized to examine the relations between lifetime Axis I disorders (other than PTSD and alcohol abuse given that PTSD and alcohol dependence were included as primary predictors) and the social network factors. These analyses revealed that Closeness was related to panic attacks [F(1, 42) = 6.902, p = .012], and social phobia [F(1, 42) = 22.725, p < .001]; Conflict was related to major depressive disorder [F(1, 42) = 4.364, p = .043] and panic attacks [F(1, 42) = 4.705, p = .036]; and Apprehension was associated with bipolar II disorder [F(1, 42) = 8.101, p = .007], generalized anxiety disorder [F(1, 42) = 5.554, p = .023], panic disorder [F(1, 42) = 6.656, p = .013], and social phobia [F(1, 42) = 33.456, p < .001]. Axis I disorders related to a social network factor were entered as covariates in the respective analysis.

3.3. Primary analyses Table 3 includes details regarding the results of group comparisons. Contrary to hypotheses, the omnibus test did not reveal significant group differences for Closeness [F(3, 40) = 2.594, p = .066]. Planned contrasts revealed that only the comorbid PTSD/AD group differed significantly from the No Axis I Diagnosis group, with the comorbid group reporting being open with and relying on friends less than those in the No Axis I Diagnosis group. The omnibus ANOVA suggested significant group differences in Conflict [F(3, 40) = 37.066, p < .001]. Both unimorbid groups and the comorbid PTSD/AD group endorsed significantly more Conflict than those in the No Axis I Diagnosis group. There were no significant differences between the unimorbid groups and the comorbid group. A significant omnibus ANOVA also emerged in relation to Family Support [F(3, 40) = 21.299, p < .001]. Those with either PTSD, AD, or comorbid PTSD/AD reported significantly less Family Support compared to those in the No Axis I Diagnosis group. Furthermore, those with comorbid PTSD/AD reported significantly less Family Support compared to those with PTSD or AD only. There also were significant group differences in Apprehension [F(3, 40) = 104.487, p < .001]. Each of the diagnostic groups endorsed significantly greater Apprehension about being close to others. Importantly, the comorbid PTSD/AD group endorsed significantly greater Apprehension than those with either PTSD or AD.2

3.4. Secondary analyses Table 4 delineates the results of the secondary ANCOVAs. In contrast to predictions, the omnibus test revealed no significant group

2 Given the low alpha for the apprehension factor, we also analyzed each of the two items separately. The same pattern emerged in these analyses as in the analysis examining the two-item Apprehension factor. Specifically, for both items the unimorbid groups reported significantly greater apprehension than the No Axis I Diagnosis comparison group and the comorbid group reported significantly greater apprehension than each of the unimorbid groups (all p’s < .01).

C.E. Dutton et al. / Journal of Anxiety Disorders 28 (2014) 67–74

71

Table 3 Means, standard errors, confidence intervals, and contrasts for social network factors without covariates. Factor

M (SE)

95% CI

Comparison to No Axis I Diagnosis

Closeness (F(3, 40) = 2.594, p = .066) 2.27 (.024) No Axis 1 Diagnosis 2.36 (.042) PTSD 2.37 (.043) AD 2.57 (1.08) PTSD/AD

(2.23, 2.32) (2.27, 2.44) (2.28, 2.46) (2.35, 2.79)

F(1, 42) = 2.50 F(1, 42) = 3.32 F(1, 42) = 6.71*

Conflict (F(3, 40) = 37.066, p < .001) No Axis 1 Diagnosis 1.26 (.014) 1.56 (.035) PTSD 1.52 (.029) AD PTSD/AD 1.64 (.075)

(1.23, 1.29) (1.49, 1.64) (1.46, 1.58) (1.49, 1.79)

F(1, 42) = 53.21** F(1, 42) = 65.33** F(1, 42) = 21.41**

Family Support (F(3, 40) = 21.299, p < .001) No Axis I Diagnosis 1.98 (.034) 2.21 (.055) PTSD AD 2.23 (.059) PTSD/AD 2.48 (.084)

(1.92, 2.05) (2.10, 2.32) (2.11, 2.35) (2.31, 2.65)

F(1, 42) = 16.31** F(1, 42) = 16.91** F(1, 42) = 25.51**

Apprehension (F(3, 40) = 104.487, p < .001) 1.33 (.017) No Axis I Diagnosis 1.88 (.066) PTSD AD 1.86 (.050) PTSD/AD 2.31 (.097)

(1.30, 1.37) (1.74, 2.01) (1.76, 1.96) (2.11, 2.50)

F(1, 42) = 70.17** F(1, 42) = 95.11** F(1, 42) = 99.75**

Comparison to comorbid group

Summary of findings

F(1, 42) = 6.71* F(1, 42) = 3.82 F(1, 42) = 3.27

PTSD/AD > No Axis I Diagnosis

F(1, 42) = 21.41** F(1, 42) = 0.73 F(1, 42) = 2.27

PTSD/AD, PTSD, AD > No Axis I Diagnosis

F(1, 42) = 25.51** F(1, 42) = 5.85* F(1, 42) = 4.55*

PTSD/AD > PTSD, AD > No Axis I Diagnosis

F(1, 42) = 99.75** F(1, 42) = 13.39** F(1, 42) = 14.92**

PTSD/AD > PTSD, AD > No Axis I Diagnosis

Note: PTSD = lifetime posttraumatic stress disorder; AD = lifetime alcohol dependence. * p ≤ .05. ** p ≤ .001.

differences in Closeness [F(3, 40) = 2.61, p = .064]. Planned contrasts confirmed this given that all group differences were nonsignificant. An omnibus ANCOVA was suggestive of group differences in relation to Conflict [F(3, 40) = 4.313, p = .010] when covarying for age, annual income, sex, panic attacks, and major depressive disorder. Both unimorbid groups and the comorbid group reported significantly more Conflict than the No Axis I Diagnosis group. The omnibus ANCOVA examining Family Support remained significant after including age, annual income, and sex as covariates [F(3, 40) = 16.051 p < .001]. Similar to the ANOVA results, those with

PTSD, AD, or comorbid PTSD/AD reported significantly less support from family than those in the No Axis I Diagnosis group. Those with comorbid PTSD/AD endorsed significantly less Family Support than those with AD or No Axis I Diagnosis. After covarying for age, annual income, sex, bipolar II disorder, generalized anxiety disorder, panic disorder, and social phobia, significant group differences in Apprehension [F(3, 40) = 18.786, p < .001] revealed that those meeting criteria for PTSD, AD, and comorbid PTSD/AD endorsed more Apprehension about utilizing social support than those in the No Axis I Diagnosis group.

Table 4 Means, standard errors, confidence intervals, and contrasts for social network factors when including covariates. Comparison to comorbid group

Summary of findings

Closeness: age, sex, income, panic attack, social phobia (F(3, 40) = 2.61, p = .064) 2.28 (.028) (2.22, 2.34) No Axis I Diagnosis 2.40 (.054) (2.29, 2.51) F(1, 42) = 3.19 PTSD 2.26 (.043) (2.18, 2.35) F(1, 42) = 0.91 AD 2.49 (.117) (2.25, 2.72) F(1, 42) = 2.60 PTSD/AD

F(1, 42) = 2.60 F(1, 42) = 0.55 F(1, 42) = 3.73

No significant group differences

Conflict: age, sex, income, major depressive disorder, panic attack (F(3, 40) = 4.313, p = .010) 1.30 (.013) (1.27, 1.33) No Axis I Diagnosis PTSD 1.42 (.034) (1.35, 1.49) F(1, 42) = 8.56** 1.42 (.036) (1.35, 1.49) F(1, 42) = 8.65** AD 1.48 (.079) (1.32, 1.64) F(1, 42) = 4.18* PTSD/AD

F(1, 42) = 4.18* F(1, 42) = 0.53 F(1, 42) = 0.63

PTSD/AD, PTSD, AD > No Axis I Diagnosis

Family Support: age, sex, income (F(3, 40) = 16.051 p < .001) 1.99 (.034) (1.92, 2.06) No Axis I Diagnosis PTSD 2.26 (.050) (2.15, 2.36) 2.12 (.051) (2.01, 2.22) AD PTSD/AD 2.46 (.084) (2.29, 2.63)

F(1, 42) = 22.80*** F(1, 42) = 3.74 F(1, 42) = 9.90**

PTSD/AD, PTSD, AD > No Axis I Diagnosis PTSD/AD > AD

Factor: covariates

M (SE)

95% CI

Comparison to No Axis I Diagnosis

F(1, 42) = 28.66*** F(1, 42) = 5.14* F(1, 42) = 22.80***

Apprehension: age, sex, income, bipolar II disorder, generalized anxiety disorder, panic disorder, social phobia (F(3, 40) = 18.786, p < .001) 1.39 (.017) (1.35, 1.42) F(1, 42) = 19.60*** No Axis I Diagnosis PTSD 1.67 (.057) (1.56, 1.79) F(1, 42) = 24.89*** F(1, 42) = 4.87* AD 1.70 (.048) (1.60, 1.80) F(1, 42) = 31.92*** F(1, 42) = 4.11* PTSD/AD 1.95 (.126) (1.70, 2.21) F(1, 42) = 19.60*** Note: PTSD = lifetime posttraumatic stress disorder; AD = lifetime alcohol dependence. * p ≤ .05. ** p ≤ .01. *** p ≤ .001.

PTSD/AD > PTSD, AD > No Axis I Diagnosis

72

C.E. Dutton et al. / Journal of Anxiety Disorders 28 (2014) 67–74

Additionally, those with comorbid PTSD/AD endorsed significantly more Apprehension than both unimorbid groups and the No Axis I Diagnosis group.3 4. Discussion Problems with social networks have been linked to both PTSD and alcohol use disorders (Brewin et al., 2000; Kaniasty & Norris, 2008; Taft, Monson, et al., 2009; Taft, Weatherill, et al., 2009; Taft et al., 2011). Unfortunately, there is little data describing social networks among people with comorbid PTSD/AD. Results were partially consistent with the general hypothesis that those with comorbid PTSD/AD would report greater problems with their social network than those with only one diagnosis, all of whom would report greater problems than those with No Axis I Diagnosis. People with comorbid PTSD/AD reported greater Apprehension about utilizing social support than people with only one or neither of these diagnoses. People with comorbid PTSD/AD also reported greater problems with Family Support than those with either PTSD, AD, or No Axis I Diagnosis. Results regarding Closeness and Conflict suggested that people with comorbid PTSD/AD experience greater problems in these domains than people No Axis I Diagnosis, but the comorbidity does not seem to be associated with elevated problems relative to the presence of only one diagnosis. People with comorbid PTSD/AD appear to experience particularly elevated problems with Family Support and Apprehension about utilizing social support. This significantly extends research suggesting that either PTSD or AD is associated with social support deficits (Beattie & Longabaugh, 1999; Brewin et al., 2000; Gordon & Zrull, 1991; Kaniasty & Norris, 2008). Given relatively little research in this domain, inferences regarding why the comorbid group reported less family support and greater apprehension are necessarily speculative. It is possible that the presence of both PTSD and AD uniquely contribute to perceptions of reduced family support and elevated apprehension about utilizing social support, and these effects are cumulative. Alternatively, the presence of one of the conditions may exacerbate the effects of the other. For example, given evidence that avoidance symptoms of PTSD are linked to relationship difficulties (Evans, McHugh, Hopwood, & Watt, 2003; Solomon, Dekel, & Mikulincer, 2008), alcohol use as an avoidance strategy may strengthen the association between PTSD avoidance symptoms and reduced social support. However, the current results leave multiple critical questions unaddressed. First, these cross-sectional data cannot speak to temporal patterning. It is possible that people who have relatively little family support and

3 As described in Section 2, analyses employed lifetime diagnoses. This decision was based on two considerations. First, despite the fact that people meeting lifetime criteria for a disorder may have no longer met criteria for PTSD or AD when social network problems were measured, the effects of PTSD and AD on social networks likely persist beyond disorder remission (O’Farrell, Murphy, Neavins, & Van Hutton, 2000; Schnurr, Hayes, Lunney, McFall, & Uddo, 2006). As such, examining links between social network problems and lifetime diagnoses yields important information for understanding associations among PTSD, AD, and social network problems. Second, the sample of people meeting criteria for lifetime comorbid PTSD/AD (complex samples estimate of n = 56.4) was notably larger than the sample meeting 12-month criteria (complex samples estimate of n = 11.4). Therefore we deemed examining lifetime diagnoses as optimal. That being said, we also tested the current hypotheses among people meeting 12-month criteria for PTSD and AD. These analyses revealed a nearly identical pattern of results. There was a significant difference on Closeness ratings among individuals who met 12-month criteria for PTSD and those with No Axis I Diagnosis in the past 12 months; those with comorbid PTSD/AD, PTSD, and AD endorsed significantly more Conflict in relationships than those with No Axis I Diagnosis; individuals who met criteria for PTSD or AD reported significantly less Family Support than those with No Axis I Diagnosis; individuals with comorbid PTSD/AD, PTSD, or AD endorsed more Apprehension about utilizing their social support network than those with No Axis I Diagnosis, and those with PTSD/AD endorsed even more Apprehension than those with a unimorbid diagnosis.

are apprehensive to enlist social support are more likely to develop PTSD and AD. Indeed, a relatively low level of social support is a risk factor for PTSD development (Soet, Brack, & Dilorio, 2003). Second, the current correlational results cannot speak to why the associations observed here emerged. For instance, Monson and colleagues (2009) hypothesized that the beliefs of significant others may influence how an individual with PTSD appraises a traumatic event, thereby influencing the course of symptoms. Future research should extend the current work by examining causal processes that may result in the links observed here (e.g., how traumatic event appraisals relate to social support, PTSD, and AD). The patterns regarding family support and apprehension are particularly interesting in light of the observations that while comorbid PTSD/AD is associated with greater social conflict and lower closeness with others compared to people without an Axis I diagnosis, the combination does not increase these problems compared to people suffering from only one of these disorders. It is possible that PTSD/AD comorbidity does not negatively impact social networks generally, but effects specific aspects of social networks. Both PTSD and AD are strongly associated with social conflict (Galovski & Lyons, 2004; Monson et al., 2009; O’Farrell & Murphy, 1995; Taft et al., 2011). It is possible that additional complications introduced by comorbidity have a negligible impact on conflict because significant conflict is already present given only one disorder. It is also possible that the mechanisms by which PTSD or AD impact conflict are significantly overlapping and, as such, the combination of these two disorders does not have an additive effect on conflict. The pattern of results also suggests that neither AD nor PTSD exacerbate the effects of the other on social conflict. Given the comparable pattern of findings, similar postulations can be made regarding closeness with others. Importantly, these findings will need to be replicated to increase confidence that PTSD/AD comorbidity does not exacerbate the social conflict and perceptions of relatively low closeness to others observed among people with either PTSD or AD. While this study provides an initial inquiry into the relation between social functioning and comorbid PTSD/AD, there are a number of limitations to be considered. In addition to the limitations discussed above (i.e., cross-sectional, correlational design), the current study relied on self-report measures. While this is a useful approach for examining perception of social networks, it does not provide information concerning objective aspects of an individual’s social network. The statistical covariance approach utilized to attempt to understand the uniqueness of the link between PTSD, AD, and social functioning is limited. Future work examining these relations should employ more strict methods (e.g., exclusion criteria to control for possible group differences) to adequately examine these issues. Also, to the best of our knowledge this is the first attempt to analyze social support in the NCS-R using factors on the social network scale. Additional examination of these factors will increase confidence in the construct validity of these factors. The aforementioned limitations notwithstanding, the current findings provide a significant contribution to our understanding of how social networks relate to comorbid PTSD/AD. The involvement of social support in treatment has been linked to positive treatment outcomes for both PTSD (Price, Gros, Strachan, Ruggiero, & Acierno, 2013) and AD (Bowers & Al-Redha, 1990; McCrady, Stout, Noel, Abrams, & Nelson, 1991). Given the importance of social networks in the treatment of PTSD and AD, much more research is now needed to describe and understand problems with social networks among people with comorbid PTSD/AD. For example, future studies should investigate the effects of treating comorbid PTSD/AD on social networks to begin to examine if this type of comorbidity maintains problems with social networks. Conversely, targeting social functioning among people with comorbid PTSD/AD will begin to elucidate the effects of social networks on the maintenance of

C.E. Dutton et al. / Journal of Anxiety Disorders 28 (2014) 67–74

this comorbidity. Furthermore, prevention programs focused on social networks, with attention to apprehension about enlisting social support and family interactions specifically, could be implemented to experimentally test the role of social networks in the development of comorbid PTSD/AD. The implications of the impaired social networks among people with PTSD and AD observed here extend beyond those relevant to PTSD and AD. House, Landis, and Umberson (1988) found the absence of social support networks (e.g., marriage, family, group affiliation) was linked with higher mortality. Indeed, social isolation has been linked to cardiovascular disease, suicides, and unintentional injury (Heaney & Israel, 2002; House et al., 1988), while social support is linked to decreased morbidity (Heaney & Israel, 2002). Beyond the public health impact seen with each of these disorders individually, the co-occurrence of PTSD and SUDs present a major public health concern given the negative physical and mental health consequences, as well as poor treatment outcomes, associated with the comorbidity (Hien et al., 2010). Given problems with social networks are also linked to negative health outcomes and individuals with comorbid PTSD and alcohol use problems are a particularly difficult population to treat, it is important to continue advancing our understanding of the interplay of each of these factors in order to develop treatments that specifically and optimally target this complex constellation of problems.

References American Psychiatric Association. (1994). Diagnostic and statistical manual of mental health disorders (4th ed.). Washington, DC: Author. Back, S. E., Jackson, J. L., Sonne, S. C., & Brady, K. T. (2005). Alcohol dependence and posttraumatic stress disorder: differences in clinical presentation and response to cognitive-behavioral therapy by order of onset. Journal of Substance Abuse and Treatment, 29, 29–37. http://dx.doi.org/10.1016/j.jsat.2005.03.002 Beattie, M. C., & Longabaugh, R. (1997). Interpersonal factors and posttreatment drinking and subjective wellbeing. Addiction, 92, 1507–1521. http://dx.doi.org/10.1111/j.1360-0443.1997.tb02871.x Beattie, M. C., & Longabaugh, R. (1999). General and alcohol-specific support following treatment. Addictive Behaviors, 24, 593–606. http://dx.doi.org/ 10.1016/S0306-4603(98)00120-8 Bonin, M. F., Norton, G. R., Asmundson, G. J., Dicurzio, S., & Pidlubney, S. (2000). Drinking away the hurt: the nature and prevalence of PTSD in substance abuse patients attending a community-based treatment program. Journal of Behavior Therapy and Experimental Psychiatry, 31, 55–66. http://dx.doi.org/10.1016/S0005-7916(00)00008-2 Bowers, T., & Al-Redha, M. (1990). A comparison of outcome with group/marital and standard/individual therapies with alcoholics. Journal of Studies on Alcohol, 51, 301–309. Retrieved from http://www.jsad.com Brewin, C. R., Andrews, B., & Valentine, J. D. (2000). Meta-analysis of risk factors for posttraumatic stress disorder in trauma-exposed adults. Journal of Consulting and Clinical Psychology, 68, 748–766. http://dx.doi.org/ 10.1037//0022-006X.68.5.748 Brown, P. J., Recupero, P. R., & Stout, R. (1995). PTSD substance abuse comorbidity and treatment utilization. Addictive Behaviors, 20, 251–254. http://dx.doi.org/ 10.1016/0306-4603(94)00060-3 Brown, P. J., & Wolfe, J. (1994). Substance abuse and post-traumatic stress disorder comorbidity. Drug and Alcohol Dependence, 35, 51–59. http://dx.doi.org/ 10.1016/0376-8716(94)90110-4 Danner, G., & Radnitz, C. (2000). Protective factors and posttraumatic stress disorder in Veterans with spinal cord injury. International Journal of Rehabilitation and Health, 5, 195–203. http://dx.doi.org/10.1023/A:1012903406285 Drapkin, M. L., Yusko, D., Yasinski, C., Oslin, D., Hembree, E. A., & Foa, E. B. (2012). Baseline functioning among individuals with posttraumatic stress disorder and alcohol dependence. Journal of Substance Abuse Treatment, 41, 186–192. http://dx.doi.org/10.1016/j.jsat.2011.02.012 Eriksson, C., Vande Kemp, H., Gorsuch, R., Hoke, S., & Foy, D. (2001). Trauma exposure and PTSD symptoms in international relief and development personnel. Journal of Traumatic Stress, 14, 205–212. http://dx.doi.org/10.1023/A:1007804119319 Evans, L., McHugh, T., Hopwood, M., & Watt, C. (2003). Chronic posttraumatic stress disorder and family functioning of Vietnam veterans and their partners. Australian and New Zealand Journal of Psychiatry, 37, 765–772. http://dx.doi.org/ 10.1111/j.1440-1614.2003.01267.x Galovski, T., & Lyons, J. A. (2004). Psychological sequelae of combat violence: a review of the impact of PTSD on the veteran’s family and possible interventions. Aggression and Violent Behavior. 9, 477–501. http://dx.doi.org/ 10.1016/S1359-1789(03)00045-4

73

Gordon, A. J., & Zrull, M. (1991). Social networks and recovery: one year after inpatient treatment. Journal of Substance Abuse Treatment, 8, 143–152. http://dx.doi.org/10.1016/0740-5472(91)90005-U Hasin, D. S., Stinson, F. S., Ogburn, E., & Grant, B. F. (2007). Prevalence, correlates, disability, and comorbidity of DSM-IV alcohol abuse and dependence in the United States. Archives of General Psychiatry, 64, 830–842. http://dx.doi.org/10.1001/archpsyc.64.7.830 House, J. S., Umberson, D., & Landis, K. R. (1988). Structures and processes of social support. Annual Reviews of Sociology, 14, 293–318. Retrieved from https://www.annualreviews.org Heaney, C. A., & Israel, B. A. (2002). Social networks and social support. In: K. Glanz, B. K. Rimer, & F. M. Lewis (Eds.), Health behavior and health education: theory research and practice (3rd ed., pp. 185–209). San Francisco, CA: Jossey-Bass. Hien, D. A., Jiang, H., Campbell, A. N. C., Hu, M., Miele, G. M., Cohen, L. R., et al. (2010). Do treatment improvements in PTSD severity affect substance use outcomes? A secondary analysis from a randomized clinical trial in NIDA’s clinical trials network. American Journal of Psychiatry, 167, 95–101. Retrieved from http://ajp.psychiatryonline.org Jacobsen, L. K., Southwick, S. M., & Kosten, T. R. (2001). Substance use disorders in patients with posttraumatic stress disorder: a review of the literature. American Journal of Psychiatry, 158, 1184–1190. http://dx.doi.org/ 10.1176/appi.ajp.158.8.1184 Kachadourian, L. K., Taft, C. T., O’Farrell, T. J., Doron-LaMarca, S., & Murphy, C. M. (2012). Correlates of intimate partner psychological aggression perpetration in a clinical sample of alcoholic men. Journal of Family Psychology, 26, 206–214. http://dx.doi.org/10.1037/a0027436 Kaniasty, K., & Norris, F. (2008). Longitudinal linkages between perceived social support and posttraumatic stress symptoms: sequential roles of social causation and social selection. Journal of Traumatic Stress, 21, 274–281. http://dx.doi.org/10.1002/jts.20334 Kessler, R. C. (2000). Posttraumatic stress disorder: the burden to the individual and to society. Journal of Clinical Psychiatry, 61, 4–12. Retrieved from https://www.psychiatrist.com Kessler, R. C., Berglund, P., Chiu, W. T., Demler, O., Heeringa, S., Hiripi, E., et al. (2004). The US National Comorbidity Survey Replication (NCS-R): design and field procedures. International Journal of Methods in Psychiatric Research, 13, 69–92. http://dx.doi.org/10.1002/mpr.167 Kessler, R. C., Crum, R. M., Warner, L. A., Nelson, C. B., Schulenber, J., & Anthony, J. C. (1997). Lifetime co-occurrence of DSM-III-R alcohol abuse and dependence with other psychiatric disorders in the National Comorbidity Survey. Archives of General Psychiatry, 54, 313–321. http://dx.doi.org/10.1001/archpsyc.1997.01830160031005 Kessler, R. C., & Merikangas, K. R. (2004). The National Comorbidity Survey Replication (NCS-R): background and aims. International Journal of Methods in Psychiatric Research, 13, 60–68. http://dx.doi.org/10.1002/mpr.166 Kessler, R. C., Sonnega, A., Bromet, E., Hughes, M., & Nelson, C. B. (1995). Posttraumatic stress disorder in the National Comorbidity Survey. Archives of General Psychiatry, 52, 1048–1060. http://dx.doi.org/ 10.1001/archpsyc.1995.03950240066012 Kessler, R. C., & Üstün, T. B. (2004). The World Mental Health (WMH) survey initiative version of the World Health Organization (WHO) Composite International Diagnostic Interview (CIDI). International Journal of Methods in Psychiatric Research, 13, 93–121. http://dx.doi.org/10.1002/mpr.168 Leonard, K. E., & Blane, H. T. (1992). Alcohol and marital aggression in a national sample of young men. Journal of Interpersonal Violence, 7, 19–30. http://dx.doi.org/10.1177/088626092007001002 Marlatt, G. A., & Gordon, J. R. (1980). Determinants of relapse: implications for the maintenance of behavior change. In: P. O. Davidson, & S. M. Davidson (Eds.), Behavioral medicine: changing health lifestyles (pp. 410–452). New York: Brunner/Mazel. Marshal, M. P. (2003). For better or for worse? The effects of alcohol use on marital functioning. Clinical Psychology Review, 23, 959–997. http://dx.doi.org/10.1016/j.cpr.2003.09.002 McCrady, B. S., Stout, R., Noel, N., Abrams, D., & Nelson, H. F. (1991). Effectiveness of three types of spouse-involved behavioral alcoholism treatment. British Journal of Addiction, 86, 1415–1424. http://dx.doi.org/10.1111/ j.1360-0443.1991.tb01727.x Miller, G., & Chapman, J. (2001). Misunderstanding analysis of covariance. Journal of Abnormal Psychology, 110, 40–48. http://dx.doi.org/10.1037/ 0021-843X.110.1.40 Mills, K., Teeson, M., Ross, J., & Peters, L. (2006). Trauma, PTSD, and substance use disorders: findings from the Australian National Survey of mental health and well-being. American Journal of Psychiatry, 163, 651–658. http://dx.doi.org/10.1176/appi.ajp.163.4.652 Monson, C. M., Taft, C. T., & Fredman, S. J. (2009). Military-related PTSD and intimate relationship: from description to theory-driven research and intervention development. Clinical Psychology Review, 29, 707–714. http://dx.doi.org/10.1016/j.cpr.2009.09.002 Monson, C. M., Fredman, S. J., Macdonald, A., Pukay-Martin, N. D., Resick, P. A., & Schnurr, P. P. (2012). Effect of cognitive-behavioral couple therapy for PTSD: a randomized controlled trial. Journal of the American Medical Association, 308, 700–709. http://dx.doi.org/10.1001/jama.2012.9307 Murphy, C. M., & O’Farrell, T. J. (1994). Factors associated with marital aggression in male alcoholics. Journal of Family Psychology, 8, 321–335. http://dx.doi.org/10.1037/0893-3200.8.3.321

74

C.E. Dutton et al. / Journal of Anxiety Disorders 28 (2014) 67–74

National Comorbidity Survey. (n.d.). NCS-R interview schedule and respondent booklet. National Comorbidity Survey. Retrieved from http://www.hcp. med.harvard.edu/ncs/replication.php Najavits, L. M., Gasfriend, D. R., Barber, J., Reif, S., Muenz, L. R., Blaine, J., et al. (1998). Cocaine dependence with and without posttraumatic stress disorder in the NIDA Cocaine Collaborative Study. American Journal of Psychiatry, 155, 214–219. Retrieved from http://ajp.psychiatryonline.org/ Najavits, L. M., Weiss, R. D., & Shaw, S. R. (1999). A clinical profile of women with posttraumatic stress disorder and substance dependence. Psychology of Addictive Behaviors, 13, 98–104. http://dx.doi.org/10.1037/0893-164X.13.2.98 O’Farrell, T. J., & Murphy, C. M. (1995). Marital violence before and after alcoholism treatment. Journal of Consulting and Clinical Psychology, 63, 256–262. http://dx.doi.org/10.1037/0022-006X.63.2.256 O’Farrell, T. J., Murphy, C. M., Neavins, T. M., & Van Hutton, V. (2000). Verbal aggression among male alcoholic patients and their wives in the year before and two years after alcoholism treatment. Journal of Family Violence, 15, 295–310. http://dx.doi.org/10.1023/A:1007503411845 Ouimette, P. C., Brown, P. J., & Najavits, L. M. (1998). Course and treatment of patients with both substance use and posttraumatic stress disorders. Addictive Behaviors, 23, 785–795. http://dx.doi.org/10.1016/S0306-4603(98)00064-1 Ouimette, P. C., Finney, J. W., & Moos, R. H. (1999). Two-year posttreatment functioning and coping of substance abuse patients with posttraumatic stress disorder. Psychology of Addictive Behaviors, 13, 105–114. http://dx.doi.org/10.1037/0893-164X.13.2.105 Ouimette, P. C., Goodwin, E., & Brown, P. J. (2006). Health and well being of substance use disorder patients with and without posttraumatic stress disorder. Addictive Behaviors, 31, 1415–1423. http://dx.doi.org/10.1016/j.addbeh.2005.11.010 Pietrzak, R. H., Goldstein, R. B., Southwick, S. M., & Grant, B. F. (2011). Prevalence and Axis I comorbidity of full and partial posttraumatic stress disorder in the United States: results from Wave 2 of the National Epidemiologic Survey on Alcohol and Related Conditions. Journal of Anxiety Disorders, 25, 456–465. http://dx.doi.org/10.1016/j.janxdis.2010.11.010 Price, M., Gros, D. F., Strachan, M., Ruggiero, K. J., & Acierno, R. (2013). The role of social support in exposure therapy for operation Iraqi freedom/operation enduring freedom veterans: a preliminary investigation. Psychological Trauma: Theory, Research, Practice and Policy, 5, 93–100. http://dx.doi.org/10.1037/a0026244 Raykov, T. (1997). Scale reliability, Cronbach’s coefficient alpha, and violations of essential tau-equivalence with fixed congeneric components. Multivariate Behavioral Research, 32, 329–353. http://dx.doi.org/10.1207/ s15327906mbr3204 2 Rash, C. J., Coffey, S. F., Baschnagel, J. S., Drobes, D. J., & Saladin, M. E. (2008). Psychometric properties of the IES-R in traumatized substance dependent individuals with and without PTSD. Addictive Behaviors, 33, 1039–1047. http://dx.doi.org/10.1016/j.addbeh.2008.04.006 Read, J. P., Brown, P., & Kahler, C. W. (2004). Substance abuse and posttraumatic stress disorders: symptom interplay and effects on outcome. Addictive Behaviors, 29, 2665–2672. http://dx.doi.org/10.1016/j.addbeh.2004.02.061

Samokhvalov, A. V., Popova, S., Room, R., Ramos, M., & Rehm, J. (2010). Disability associated with alcohol abuse and dependence. Alcoholism, Clinical and Experimental Research, 34, 1871–1878. http://dx.doi.org/ 10.1111/j.1530-0277.2010.01275.x Schnurr, P. P., Hayes, A. F., Lunney, C. A., McFall, M., & Uddo, M. (2006). Longitudinal analysis of the relationship between symptoms and quality of life in veterans treated for posttraumatic stress disorder. Journal of Consulting and Clinical Psychology, 74, 707–713. http://dx.doi.org/10.1037/0022-006X.74.4.707 Soet, J. E., Brack, G. A., & Dilorio, C. (2003). Prevalence and predictors of women’s experience of psychological trauma during childbirth. Birth: Issues in Perinatal Care, 30, 36–46. http://dx.doi.org/10.1046/j.1523-536X.2003.00215.x Solomon, Z., Dekel, R., & Mikulincer, M. (2008). Complex trauma of war captivity: a prospective study of attachment and post-traumatic stress disorder. Psychological Medicine, 38, 1427–1434. http://dx.doi.org/10.1017/S0033291708002808 Stewart, S. H. (1996). Alcohol abuse in individuals exposed to trauma: a critical review. Psychological Bulletin, 120, 83–112. http://dx.doi.org/ 10.1037/0033-2909.120.1.83 Taft, C. T., Monson, C. M., Schumm, J. A., Watkins, L. E., Panuzio, J., & Resick, P. A. (2009). Posttraumatic stress disorder symptoms, relationship adjustment and relationship aggression in a sample of female flood victims. Journal of Family Violence, 24, 389–396. http://dx.doi.org/10.1007/s10896-009-9241-8 Taft, C. T., Watkins, L. E., Stafford, J., Street, A. E., & Monson, C. M. (2011). Posttraumatic stress disorder and intimate relationship problems: a meta-analysis. Journal of Consulting and Clinical Psychology, 79, 22–33. http://dx.doi.org/10.1037/a0022196 Taft, C. T., Weatherill, R. P., Woodward, H. E., Pinto, L. A., Watkins, L. E., Miller, M. W., & Dekel, R. (2009). Intimate partner and general aggression perpetration among combat veterans presenting to a posttraumatic stress disorder clinic. American Journal of Orthopsychiatry, 79, 461–468. http://dx.doi.org/10.1037/a0016657 Tucker, P., Pfefferbaum, B., Nixon, S., & Dickson, W. (2000). Predictors of posttraumatic stress symptoms in Oklahoma City: exposure, social support, peri-traumatic responses. The Journal of Behavioral Health Services & Research, 27, 406–416. http://dx.doi.org/10.1007/BF02287822 Tucker, J. A., Vuchicich, R. E., & Pukish, M. M. (1995). Molar environmental contexts surrounding recovery from alcohol problems by treatment and untreated problem drinkers. Experimental and Clinical Psychopharmacology, 3, 195–204. http://dx.doi.org/10.1037/1064-1297.3.2.195 Turner, R. J. (1981). Social support as a contingency in psychological well-being. Journal of Health and Social Behavior, 4, 357–367. http://dx.doi.org/10.2307/2136677 Zinbarg, R. E., Suzuki, S., Uliaszek, A. A., & Lewis, A. R. (2010). Biased parameter estimates and inflated type I error rates in analysis of covariance (and analysis of partial variance) arising from unreliability: alternatives and remedial strategies. Journal of Abnormal Psychology, 119, 307–319. http://dx.doi.org/10.1037/a0017552