Veterans’ PTSD Symptoms and their Partners’ Desired Changes in Key Relationship Domains

Abstract

There is a growing literature investigating the connection between veterans’ posttraumatic stress disorder (PTSD) symptoms and intimate relationship problems. Little to no work, however, has examined the connection between veterans’ PTSD symptoms and their partners’ perceptions of specific relationship areas in need of change. We examined associations between overall PTSD symptoms and symptom cluster scores with partners’ desired changes in the areas of intimacy, shared activities, and responsibilities. The sample consisted of 249 male veterans of different service eras and their female partners. Results indicated that veterans’ PTSD symptoms were associated with greater desired changes from their partners in the veterans’ intimacy behaviors and participation in shared activities. When examining the contribution of each symptom cluster individually, only the veterans’ emotional numbing symptoms emerged as a significant unique predictor and were associated with partners’ desired changes in intimacy. The findings suggest that intimacy and shared activities may be relevant areas to address in PTSD treatment for veterans and their partners, and highlight the particular significance of emotional numbing symptoms to intimacy in veterans’ relationships.

As combat veterans return home from deployment, increasing research focus has been devoted to understanding the impact of their posttraumatic stress disorder (PTSD) symptoms on their intimate relationships. A meta-analysis found a medium-sized association between PTSD symptoms and intimate relationship discord, and this association was higher for military samples than for civilian samples (Taft, Watkins, Stafford, Street, & Monson, 2011). The importance of developing a better understanding of this relationship is underscored by research suggesting that poor family functioning may, in turn, negatively influence veterans’ response to PTSD treatment (Evans, Cowlishaw, & Hopwood, 2009). However, we are aware of no study that has examined the association between veterans’ PTSD symptoms and their partners’ desires for the veterans to improve in key relationship domains. Such research is needed, as it holds particular relevance to emerging interventions that involve both veterans and partners in order to treat PTSD symptoms and relationship problems concurrently (Monson et al., 2012; Sautter, Armelie, Glynn, & Wielt, 2011). This study investigated the relationship between veterans’ PTSD symptoms and their partners’ desires for the veterans to change in three important relationship areas: intimacy, shared activities, and responsibilities.

Intimacy represents a crucial aspect of relationships, and is highly correlated with overall relationship satisfaction (Greeff, & Malherbe, 2001). Although the construct has been defined in a number of ways, most definitions indicate that trusting one another, sharing emotional and physical closeness, disclosing one’s thoughts and emotions, and being aware of one’s partner’s thoughts and emotions are all important features of intimacy (Moss & Schwebel, 1993). People who have been traumatized may have difficulty trusting, sharing with, and feeling close to their partners (Mills & Turnbull, 2004). Research suggests that certain types of trauma, such as interpersonal trauma involving betrayal, may be particularly associated with reduced trust of relationship partners (Gobin & Freyd, 2014). Among veterans, prior studies have demonstrated a connection between PTSD symptoms and self-reported marital intimacy. Zerach, Anat, Solomon, and Heruti (2010) found that deficits in marital intimacy partially mediated the relationship between PTSD symptoms and dyadic adjustment in a sample of Israeli ex-prisoners of war. The emotional numbing symptoms of PTSD, which include feelings of detachment from others and a restricted range of emotions, may be particularly involved in hindering intimacy. For example, Riggs, Byrne, Weathers, and Litz (1998) found that Vietnam veterans’ emotional numbing symptoms significantly predicted their fear of intimacy. In another study among Operation Enduring Freedom and Operation Iraqi Freedom (OEF/OIF) veterans, emotional numbing was the only PTSD symptom cluster to significantly predict problems with sexual functioning, which may be closely related to difficulties with intimacy (Nunnink, Goldwaser, Afari, Nievergelt, & Baker, 2010).

Alongside intimacy, the activities that couples share and the amount of time spent together can have a vital impact on relationship quality. In particular, couples’ participation in novel and arousing activities has been shown to increase relationship satisfaction, which occurs in part by enhancing feelings of excitement with the relationship (Aron, Norman, Aron, McKenna, & Heyman, 2000). Indirect evidence suggests that PTSD may interfere with participation in relationship activities. For example, Kuhn, Blanchard, and Hickling (2003) found that motor vehicle accident survivors with PTSD reported less participation in recreational activities than did accident survivors without PTSD. The avoidance and emotional numbing symptoms of PTSD may be most relevant to reduced participation in shared activities, as the former involves “efforts to avoid activities, places, or people that arouse recollections of the trauma,” and the latter involves “markedly diminished interest or participation in significant activities” (American Psychiatric Association [APA], 1994). Despite the importance of shared activities in intimate relationships, no study we are aware of has examined the connection between veterans’ PTSD symptoms and their participation in activities with their partners.

Long-term intimate relationships often involve a number of collective responsibilities, such as meeting financial obligations and completing housework. One potential cause of distress in veterans’ intimate relationships is the reduced ability to fulfill such responsibilities as a result of PTSD-related functional impairment. This idea has been examined indirectly in the literature on caregiver burden, which refers to caregivers’ perceptions that their emotional or physical health, social life, or financial status are affected by their caring for an impaired relative (Zarit, Todd, & Zarit, 1986). Beckham, Lytle, and Feldman (1996) found that Vietnam veterans’ PTSD symptom severity was associated with partners’ self-reported caregiver burden, and that caregiver burden was positively associated with partners’ psychological distress. In contrast, a previous study of OEF/OIF veterans did not find PTSD diagnostic status to be associated with ambiguity or conflict about family responsibilities (Sayers, Farrow, Ross, & Oslin, 2009). However, this previous study examined PTSD and family responsibilities as dichotomous variables, and thus may have had reduced power to detect a relationship. In the current study, we examine this relationship using continuous variables.

Given the importance of intimacy, shared activities, and responsibilities in relationships, the current study investigated associations between veterans’ PTSD symptoms and their partners’ desires for the veterans to change in these domains. We hypothesized that PTSD symptom severity would be associated with partners reporting greater desires for the veterans to improve in each domain. As clusters of PTSD symptoms have been shown to differentially associate with relationship and family adjustment (Monson, Taft, & Fredman, 2009), we examined symptom clusters individually alongside total PTSD symptoms. We hypothesized that the intimacy domain would show the strongest association with emotional numbing symptoms, and that the shared activities domain would show the strongest associations with avoidance symptoms and emotional numbing symptoms. We did not hypothesize the responsibilities domain to show a particularly strong association with any one symptom cluster, as each symptom type could conceivably interfere with relationship responsibilities.

Method

Participants

The study was conducted at the Veterans Affairs (VA) Boston Healthcare System and the New Mexico VA Healthcare System. Study procedures were approved and reviewed annually by the relevant Institutional Review Board at each institution, and informed consent was obtained from each participant prior to enrolling in the study. The veterans were recruited for the study via flyers, mailings, and clinician referrals. Eligibility criteria were as follows: (1) the veteran must have been living with an intimate partner for the prior 12 months, (2) the veteran must have endorsed a history of exposure to at least one traumatic event meeting the Diagnostic and Statistical Manual of Mental Disorders (4^th ed., DSM-IV; APA, 1994) PTSD Criterion A, and (3) the veteran’s partner must have agreed to participate in the study. A total of 298 couples had initially enrolled in the study. Twelve couples were excluded for the following reasons: two were deemed ineligible after being consented, three withdrew from the study voluntarily, and seven were terminated because one or both members of the couple had difficulty conforming to protocol requirements. Additionally, 9 couples included a veteran that did not complete the structured clinical interview for PTSD symptoms (Blake et al., 1990), and 28 couples in which the veteran was female were omitted from analyses, as there were not sufficient cases to investigate the influence of gender on the relationships examined. The final sample consisted of 249 male-veteran and female-partner couples, with 91.6% of the couples comprised of one veteran and one non-veteran partner, and 8.4% comprised of both veterans. In these cases, the person who initiated contact with the study staff and was screened for eligibility was labeled as the “veteran.”

The majority of the sample (82.9%) self-identified as Caucasian, 9.8% as Black or African American, 8.1% as American Indian or Alaskan Native, 1.6% as Asian, 0.6% as Hawaiian or other Pacific Islander, and 6.5% as unknown racial origin (demographic categories were not mutually exclusive). In addition, 20.2% endorsed Hispanic or Latino ethnicity. The average age of the sample was 51.9 (SD = 11.2). Of the couples in the study, 82.7% were married, 82.4% had been together 5 years or more, and 42.9% had been together for over 20 years. Veterans’ reported eras of military service were: 60.1% Vietnam War, 17.3% OEF/OIF, 12.5% Operation Desert Storm, 0.8% Korean War, and 10.5% other eras. Regarding veterans’ branch of military service, 52.8% had served in the Army, 19.8% in the Marines, 16.1% in the Navy, 13.7% in the Air Force, and 3.6% in the Coast Guard.

Measures

The Clinician Administered PTSD Scale (CAPS; Blake et al., 1990)

The CAPS is a structured diagnostic interview that evaluates the 17 PTSD symptoms as established by the DSM-IV (APA, 1994). The clinician conducting the interview rates the frequency and intensity of each symptom on a 5-point scale. The frequency and intensity scores are combined to form a severity score for each symptom, which are then combined to form individual symptom cluster severity scores and a total PTSD severity score. The avoidance and emotional numbing symptoms were disaggregated in the current study in order to examine the individual effects of each. Possible total scores on the CAPS range from 0 to 136. For the reexperiencing, emotional numbing, and hyperarousal symptom clusters, the possible score range is 0 to 40, and for the avoidance symptom cluster the possible score range is 0 to 16. Current PTSD diagnostic status was assessed using a validated scoring rule (i.e., at least one reexperiencing symptoms, three avoidance/emotional numbing symptoms, and two hyperarousal symptoms with a frequency of one or greater and an intensity of two or greater). In the sample, 42.2% of the veterans met diagnostic criteria for PTSD. Twenty-three percent of the CAPS interviews were scored at random by a second rater, and a previous study with this dataset reported high inter-rater reliability (intraclass correlation coefficient = .99; Miller et al., 2013).

Desired Changes Questionnaire (Heavey, Lane, & Christensen, 1993)

The DCQ asks participants to rate the degree to which they want their partner to change in various relationship behaviors on a scale of 1 (no change) to 7 (much more change). In this study, 17 items from the DCQ were organized into three subscales. The Intimacy subscale consisted of 5 items (i.e., be more affectionate with me, show appreciation for things I do well, accept praise, pay attention to my sexual needs, and express his emotions clearly); the Shared Activities subscale consisted of 9 items (i.e., get together with my friends, start interesting conversations with me, go out with me, get together with my relatives, get together with our friends, agree to do things I like when we go out together, spend time with me and not other men/women, spend time with me, and spend time in outside activities); and the Responsibilities subscale consisted of 3 items (i.e., assume responsibility for finances, accomplish responsibilities promptly, and participate in decisions about spending money). Coefficient α for the Intimacy, Shared Activities, and Responsibilities subscales were .77, .88, and .79, respectively.

Data Analysis

Descriptive statistics were analyzed for all study variables. Next, Pearson’s product-moment correlations were calculated between veterans’ PTSD symptom severity scores (i.e., total PTSD severity, reexperiencing severity, avoidance severity, emotional numbing severity, and hyperarousal severity) and the degree of partners’ desired changes in intimacy, shared activities, and responsibilities. Finally, we investigated the unique contribution of each significantly correlated symptom cluster in separate multiple regression analyses for desired changes in intimacy, shared activities, and responsibilities. All effect sizes were interpreted using Cohen’s (1988) recommendations for small (.10 ≤ r < .30), medium (.30 ≤ r < .50), and large (r ≥ .50) effects.

Results

Descriptive Statistics

Ranges, means, and standard deviations were calculated for all study variables. Veterans’ total current PTSD symptom severity ranged from 0 to 109 (M = 42.16, SD = 28.17), reexperiencing symptom severity ranged from 0 to 34 (M = 11.04, SD = 9.38), avoidance symptom severity ranged from 0 to 16 (M = 4.92, SD = 4.70), emotional numbing symptom severity ranged from 0 to 33 (M = 10.10, SD = 9.06), and hyperarousal symptom severity ranged from 0 to 37 (M = 16.09, SD = 9.71). Partners’ desired changes in intimacy ranged from 5 to 35 (M = 16.92, SD = 7.81), desired changes in shared activities ranged from 9 to 63 (M = 23.72, SD = 12.56), and desired changes in responsibilities ranged from 3 to 21 (M = 9.14, SD = 5.58).

Veterans’ PTSD Symptoms and Partners’ Desired Changes

The relationships between veterans’ PTSD symptom severity scores and their partners’ desired changes in intimacy, shared activities, and responsibilities were examined with bivariate correlations. Total PTSD severity scores and individual symptom cluster severity scores showed large significant correlations with one another (r = .58–.88, all p < .001). Bivariate correlations between veterans’ PTSD symptom severity scores and partners’ desired changes are displayed in Table 1. Results indicated that veterans’ symptom severities for total PTSD, reexperiencing, emotional numbing, and hyperarousal were positively correlated with their partners’ desired changes in intimacy. Additionally, total PTSD symptom severity and the severity of each symptom cluster were positively correlated with partners’ desired changes in shared activities. Finally, total PTSD symptom severity, reexperiencing symptom severity, emotional numbing symptom severity, and hyperarousal symptom severity were positively correlated with their partners’ desired changes in responsibilities.

Table 1.

Bivariate Correlations between Veterans’ PTSD Symptoms and Partners’ Desired Changes

Symptom Severity	DCQ: Intimacy	DCQ: Shared Activities	DCQ: Responsibilities
Total PTSD	.20**	.19**	.15*
Reexperiencing	.14*	.18**	.13*
Avoidance	.11	.13*	.09
Emotional Numbing	.22***	.16**	.13*
Hyperarousal	.17**	.17**	.14*

Note. Abbreviations: PTSD, posttraumatic stress disorder; DCQ, Desired Changes Questionnaire.

^*p <.05.

^**p <.01.

^***p <.001.

Table 2 presents the results of the multiple regression analyses for partners’ desired changes. For the model explaining partners’ desired changes in intimacy, veterans’ reexperiencing symptom severity, emotional numbing symptom severity, and hyperarousal symptom severity accounted for a significant portion of the variance, R² = .05, F(3, 245) = 4.63, p < .01. Veterans’ emotional numbing symptom severity emerged as the only unique predictor. For the model explaining partners’ desired changes in shared activities, veterans’ reexperiencing symptom severity, avoidance symptom severity, emotional numbing symptom severity, and hyperarousal symptom severity accounted for a significant portion of the variance, R² = .04, F(4, 244) = 2.57, p < .05, although no single symptom cluster emerged as a unique predictor. The model explaining partners’ desired changes in responsibilities was not significant, R² = .02, F(3, 244) = 1.93, ns.

Table 2.

Multiple Regression Results: Predictors of Desired Changes in Intimacy and Shared Activities

Symptom Severity	B	SE B	β	Partial r
DCQ: Intimacy
Reexperiencing	.00	.07	.00	.00
Emotional Numbing	.16	.07	.19*	.14
Hyperarousal	.04	.08	.04	.03
DCQ: Shared Activities
Reexperiencing	.18	.13	.14	.09
Avoidance	−.12	.25	−.05	−.03
Emotional Numbing	.10	.12	.08	.05
Hyperarousal	.07	.12	.06	.04

Note. Abbreviations: PTSD, posttraumatic stress disorder; DCQ, Desired Changes Questionnaire.

^*p <.05.

^**p <.01.

^***p <.001.

Discussion

The aim of this study was to examine the connection between veterans’ PTSD symptoms and their partners’ desires for the veterans to change in the key relationship domains of intimacy, shared activities, and responsibilities. Results indicated that the severity of veterans’ PTSD symptoms was correlated at the bivariate level with the degree of change desired by the partners in each domain. When investigating the contribution of each symptom cluster in regression analyses, only veterans’ emotional numbing symptoms uniquely predicted their partners’ desires for change in the veterans’ intimacy behaviors.

Findings of the current study represent a novel contribution to the literature on PTSD-related relationship problems among veterans and hold implications for emerging treatment approaches with veterans and their partners. First, although previous research has examined the connection between veterans’ PTSD and intimacy (Riggs et al., 1998; Zerach et al., 2010), we are not aware of any prior studies that have investigated PTSD in relation to partners’ desires for changes in veterans’ intimacy behaviors. While ensuring that the relationship between PTSD and intimacy is not due to potentially biased reporting from the veteran, this approach is more germane to treatments that address partners’ perceptions of the veterans’ intimacy behaviors (Monson et al., 2012; Sautter et al., 2011). The finding that veterans’ PTSD symptoms were associated with partners’ desired changes in intimacy behaviors adds to growing documentation of PTSD-related intimacy difficulties by showing that such difficulties are of particular concern to veterans’ partners. Additionally, the finding that emotional numbing was the only symptom cluster to uniquely predict partners’ desired changes in intimacy bolsters prior research showing the particular significance of this cluster with respect to relationship functioning (Cook, Riggs, Thompson, Coyne, & Sheikh, 2004; Nunnink et al., 2010; Riggs et al., 1998) and supports targeting emotional numbing symptoms to address intimacy deficits between veterans and their partners (Monson et al., 2012; Sautter et al., 2011).

Results indicated that veterans’ PTSD symptoms were associated with partners’ desired changes in shared activities, suggesting that veterans’ symptoms may restrict them from going out and spending time with their partners, and this may be an important relationship area to target in treatment. Contrary to our hypothesis, neither avoidance nor emotional numbing uniquely predicted partners’ desired changes in shared activities. In part, this may have been due to the difficulty in parsing out the individual influences of symptom clusters that are highly associated with one another. However, it is also possible that each symptom cluster plays a role in this relationship. For example, an aspect of a shared activity (e.g., loud noises during a movie) may trigger distressing reexperiencing symptoms, which may then cause the veteran to avoid similar activities. Additionally, hyperarousal symptoms (e.g., a heightened need to scan for threat cues) may make the prospect of going out seem too overwhelming, and may contribute to the diminished interest in activities captured in the emotional numbing symptoms. It also seems likely that there is a reciprocal relationship between reduced participation in shared activities and reduced intimacy between partners.

Despite the lack of a uniquely predictive symptom cluster, the findings regarding shared activities indicate that this may be an important avenue for future research. For instance, while engaging in novel and arousing activities is particularly important in enhancing relationship satisfaction (Aron et al., 2001), veterans with PTSD may be especially averse to such activities because novel situations involve interpretation of ambiguous stimuli (Lazarus, & Folkman, 1984), and veterans with PTSD may have a tendency to interpret ambiguous stimuli as hostile or threatening (Chemtob, Novaco, Hamada, Gross, & Smith, 1997).

This study is also one of the first to empirically examine the potential influence of PTSD on veterans’ relationship responsibilities. While there were bivariate associations between veterans’ PTSD symptoms and their partners’ desired changes in the responsibilities domain, the multivariate model explaining partners’ desired changes in responsibilities was not significant. This is consistent with a prior study that did not find a diagnosis of PTSD to be related to ambiguity or conflict about family responsibilities (Sayers et al., 2009). It is possible that PTSD symptoms do not impact relationship responsibilities as strongly as they impact other aspects of veterans’ relationships. However, it is important to note that in the study by Sayers and colleagues (2009), family responsibilities were only assessed using two items, and in the current study, they are only assessed using three items. Thus, future research may find an association between PTSD symptoms and the fulfillment of relationship responsibilities using a more thorough assessment of this domain.

Several limitations of the current study should be noted. First, while a number of associations were found between the variables of interest, they were small in magnitude. In part, this may have been a result of examining associations across reporters (i.e., clinician’s interview-based rating of veteran’s PTSD symptoms and partner’s self-report rating of desired changes in veteran’s behaviors), which reduces the possibility that an overall reporting bias enhances the relationship between the variables. Additionally, the assessment of partners’ desired changes in the veterans’ behavior includes more variability than would the assessment of partners’ ratings of the veterans’ behavior, as there may be differences between the partners in what behaviors they value over others. Nevertheless, we see the assessment of partners’ desired changes as a strength of the study because partners’ interpretations of and opinions about veterans’ behavior is a crucial aspect of couples therapy for PTSD (Monson & Fredman, 2012). Another limitation is that the items from the DCQ were grouped into subscales for the purposes of this study, and factor loadings were not examined. It is possible that some items may relate to more than one relationship domain (i.e. intimacy and shared activities). Moreover, the three relationship domains assessed in the current study do not encompass other factors that may represent important links between PTSD and relationship distress, such as the demand-withdraw communication pattern (Cook et al., 2004) and intimate partner violence (Taft et al., 2011).

It is also important to note that the demographic characteristics of the sample limit the ability to generalize the findings to all veterans. The sample was comprised of male-veteran female-partner couples in heterosexual relationships that were primarily long-standing. Additionally, the majority of the sample was Caucasian and had served during the Vietnam War era. The current study was also not able to investigate the potential influence of gender on the variables examined. Future research among samples with even numbers of male and female veterans and partners may find gender to moderate these associations. Finally, while the interpretation that PTSD symptoms play a causal role in relationship problems is in line with much of the literature on this phenomenon (Monson et al., 2009), the cross-sectional nature of this study limits a better understanding of the directionality of this relationship.

In conclusion, study results suggest important implications for the research and treatment of veterans’ with PTSD and their partners. Veterans’ PTSD symptoms were associated with greater desired changes from the partners in the veterans’ intimacy behaviors and participation in shared activities, suggesting that these are relevant areas of relationship functioning to address in treatment. The finding that emotional numbing symptoms uniquely predicted partners’ desired changes in the veterans’ intimacy behaviors extends support to treatment approaches that target these symptoms specifically (Monson et al., 2012; Sautter et al., 2011). Results also offer directions for future research. The development over time of relationship problems associated with veterans’ PTSD symptoms needs further elucidation, and longitudinal studies could incorporate an assessment of key relationship areas that have not received much study, such as shared activities and responsibilities. Additionally, it will be helpful for future research to identify potential moderators of the relationships between veterans’ PTSD symptoms and partners’ desired changes. For instance, Renshaw and Campbell (2011) found that partners’ perceptions of veterans’ deployment experiences moderated the association between veterans’ PTSD symptoms and partners’ psychological and relationship distress, such that their distress decreased as their perceptions increased. It seems likely that partners’ perceptions and attributions of the veterans’ symptoms would similarly moderate the associations found in the current study. Overall, the findings contribute novel evidence to a growing literature on relationship problems among veterans with PTSD, and highlight key relationship areas that may be relevant to treatment with veterans and their partners.