Understanding the relationship of perceived social support to post-trauma cognitions and posttraumatic stress disorder

Abstract

Poor social support in the aftermath of a traumatic event is a well-established risk factor for posttraumatic stress disorder (PTSD) among adult trauma survivors. Yet, a great deal about the relationship between social support and PTSD remains poorly understood. In this study, we analyzed data from 102 survivors of a serious motor vehicle accident (MVA) at 4 weeks (Time 1) and 16 weeks (Time 2) post-MVA. We assessed the role of perceived dyadic social support, positive dyadic interaction, and negative dyadic interaction in the development and maintenance of PTSD. In addition, we examined how these social support constructs work together with negative post-trauma cognitions to affect the maintenance of PTSD. Neither perceived social support nor the quality of social interaction (i.e., positive or negative) was associated with PTSD symptom severity at Time 1. However, among those with elevated PTSD symptom severity at Time 1, greater social support and positive social interaction and lower negative social interaction were each associated with reductions in PTSD symptom severity from Time 1 to Time 2. For social support and negative social interaction, this association ceased to be significant when jointly assessed with negative post-trauma cognitions, suggesting that perceived social support and negative dyadic interaction were associated with maintenance of PTSD symptom severity because of their association with negative post-trauma cognitions. These results provide support to models and treatments of PTSD that emphasize the role of negative post-trauma cognitions in maintenance of PTSD.

Introduction

Poor social support following a traumatic event is among the greatest risk factors for posttraumatic stress disorder (PTSD) across types of trauma (Brewin et al., 2000, Ozer et al., 2003), including motor vehicle accidents (Dougall, Ursano, Posluszny, Fullerton, & Baum, 2001). In early conceptualizations of social support and PTSD, theorists posited that the presence of positive social support buffers against development of post-stress psychopathology (Cohen & Wills, 1985) and facilitates the process of recovery (Burgess & Holmstrom, 1978). However, subsequent findings have painted a more nuanced picture. Researchers have found that the presence of negative social interactions (e.g., expressed emotion or interpersonal friction) may be a stronger predictor of PTSD than the absence of positive social support (Tarrier et al., 1999, Ullman, 1996, Zoellner et al., 1999). In addition, directionality of the association between social support and PTSD has been shown to vary over time. In earlier stages of coping with trauma (e.g., 6–12 months post-trauma), poor social support acts as a risk factor for greater PTSD symptom severity (Kaniasty & Norris, 2008). During later stages of coping with trauma (18–24 months post-trauma), greater PTSD severity contributes to an erosion of social support resources (Kaniasty & Norris, 2008; see also King, Taft, King, Hammond, & Stone, 2006).

Although these findings have advanced our understanding of the association between social support and PTSD, a great deal remains poorly understood. For example, a variety of method limitations in many prior studies (e.g., cross-sectional design and study time frame) has prevented researchers from clearly disentangling social support's role in the initial development of PTSD from its role in the maintenance of PTSD over the first several months or years post-trauma (for a review of social support and PTSD, see Guay, Billette, & Marchand, 2006). In studies that have utilized longitudinal designs beginning within one month after the trauma, results have been somewhat inconsistent. For example, Cook and Bickman (1990) found that social support was not associated with psychological distress (i.e., depression, anxiety, and somatization) 1 week following a natural disaster, but that this association was significant at 6, 11, and 16 weeks post-disaster. Relative to those with low levels of social support, those with high levels of support exhibited comparable levels of psychological distress at week 1 but exhibited a sharper decline in distress over time. These findings suggest that social support is associated with maintenance, but not development, of post-traumatic stress symptoms. However, due to sample size limitations, these researchers were unable to prospectively examine social support's contribution to the maintenance of elevated psychological distress.

In contrast, Zoellner et al. (1999) found that interpersonal friction, but not positive social support, was significantly associated with PTSD and depression symptoms at 2 weeks post-trauma. In addition, interpersonal friction predicted PTSD symptom severity at 3 months post-trauma beyond the effects of PTSD at 2 weeks post-trauma. Similarly, Andrews, Brewin, and Rose (2003) found that social support satisfaction and negative responses from social support figures were each associated with PTSD symptom severity at 1-month post-trauma but that only negative responses were associated with PTSD symptom severity at 6-months post-loss after controlling for 1-month PTSD symptom severity. However, in both of these studies, the course of PTSD symptoms was examined for the entire sample. Accordingly, these findings reflect the course of symptoms both for those who developed acute PTSD as well as those who did not. It is unclear whether social support (positive or negative) would have been associated with the course of PTSD symptom severity in only the subset with initially elevated PTSD symptoms.

In addition to issues regarding the development and maintenance of PTSD, little is known about the mechanisms through which social support exerts its influence on either the development or maintenance of PTSD (Vogt, King, & King, 2007). Indeed, it is unclear whether various aspects of social support (e.g., positive vs. negative) operate through the same or different mechanisms. This gap in the literature is of concern as clarifying the etiological processes by which social support and other risk factors lead to PTSD is critical to the optimal development and delivery of treatments for those suffering from the disorder (Kraemer, Stice, Kazdin, Offord, & Kupfer, 2001). Several theorists have posited that greater social support may impact PTSD by impeding the development and persistence of negative post-trauma cognitions (Ehlers and Clark, 2000, Guay et al., 2006, Joseph et al., 1997). Negative post-trauma cognitions of the self (i.e., a sense of the self as incompetent or self-blame regarding the traumatic event) and the world (i.e., belief that the world is entirely dangerous) are thought to contribute to both the development and maintenance of PTSD by producing an ongoing sense of threat (Brewin and Holmes, 2003, Ehlers and Clark, 2000, Foa and Rothbaum, 1998, Resick and Schnicke, 1992). Negative cognitions about the self and the world discriminate between traumatized individuals with and without PTSD (Beck et al., 2004, Foa et al., 1999) and prospectively predict PTSD symptom severity (Dunmore et al., 2001, Ehlers et al., 1998, Ehring et al., 2008). Accordingly, there is a strong impetus for examining the association of social support to post-trauma cognitions and their joint association with the development and maintenance of PTSD symptom severity.

In this study, we had three aims: to examine the association of social support to the (1) development and maintenance of PTSD symptoms, (2) the development and maintenance of negative post-trauma cognitions, and (3) to examine how social support and negative post-trauma cognitions work together to affect the development and maintenance of PTSD symptoms. To our knowledge, this study is the first to jointly assess post-trauma cognitions and social support as predictors of PTSD symptom severity in a longitudinal study of adult trauma survivors. To assess these aims, we examined data from 102 survivors of a serious motor vehicle accident (MVA) at 4 weeks (Time 1), and 16 weeks (Time 2) post-MVA. For each aim, we examined three social support constructs: the level of perceived dyadic social support (DSS), negative dyadic interaction (dyadic conflict; DC), and positive dyadic interaction (dyadic depth; DD). Based on previous theoretical and empirical work (Guay et al., 2006), we hypothesized that each social support construct would be associated with both the development and maintenance of PTSD symptoms. We further hypothesized that individuals with greater perceived dyadic social support, greater perceived positive dyadic interaction, and less perceived negative dyadic interaction would report less severe initial negative post-trauma cognitions as well as a reduction in negative post-trauma cognitions over time. Finally, we hypothesized that when jointly assessed with post-trauma cognitions these social support constructs would no longer be associated with PTSD symptom development or maintenance.