Research report
Co-occurring posttraumatic stress and depression symptoms after sexual assault: A latent profile analysis

https://doi.org/10.1016/j.jad.2013.01.026Get rights and content

Abstract

Background

Symptoms of posttraumatic stress disorder (PTSD) and depression frequently co-occur, but their distinctiveness following trauma remains unclear. We examined patterns of PTSD and depression symptoms after sexual assault to evaluate the extent to which assault survivors primarily reported symptoms of both disorders or whether there were meaningfully distinct subgroups with discordant PTSD and depression symptoms.

Methods

Latent profile analysis was used to examine self-reported PTSD and depression symptoms among 119 female sexual assault survivors at 1-, 2-, 3-, and 4-months post-assault.

Results

At all time points, a 4-class solution fit the data best, revealing four subgroups with low, low-moderate, high-moderate, and severe levels of both PTSD and depression symptoms. Within each subgroup, PTSD symptom severity co-occurred with comparable depression symptom severity. At no time point were there reliable subgroups with discordant PTSD and depression symptom severities. Emotional numbing, hyperarousal, and overall PTSD symptom severity reliably distinguished each class from the others. Class membership at 1-month post-assault predicted subsequent class membership and functional impairment.

Limitations

Additional research is needed to evaluate predictors of class membership, temporal stability of classes, and generalizability to other trauma populations.

Conclusions

Co-occurring and comparably severe PTSD and depression symptoms are pervasive among female sexual assault survivors. The absence of a distinct subset of individuals with only PTSD or depression symptoms suggests that PTSD and depression may be manifestations of a general posttraumatic stress response rather than distinct disorders after trauma. Integrated treatments targeting both PTSD and depression symptoms may therefore prove more efficient and effective.

Introduction

The majority of individuals experience one or more potentially traumatizing events in their lifetime (Kessler et al., 1995). While most trauma survivors maintain healthy functioning or gradually recover (Bonanno, 2004), depending on the nature of the event, approximately 8% to 54% experience long-term distress, impairment, and mental disorders such as posttraumatic stress disorder (PTSD; Breslau et al., 1998, Bryant et al., 2010, Kessler et al., 1995). Although PTSD is the most commonly researched trauma disorder, major depression occurs at a similar rate after trauma and frequently co-occurs with PTSD (Bryant et al., 2010, Shalev et al., 1998).

Clarifying the relationship between PTSD and depression in the aftermath of trauma has important conceptual, diagnostic, and treatment implications. Some researchers have proposed that PTSD and depression co-occur at high rates after trauma because they are manifestations of a single, underlying posttraumatic psychopathology (Norman et al., 2011, O’Donnell et al., 2004). Supporting this hypothesis, studies have found that PTSD and post-traumatic depression share nearly identical risk factors and follow a similar time-course (Brewin et al., 2000, Bromet et al., 1998, deRoon-Cassini et al., 2010, Kendler et al., 2002, Norman et al., 2011, O’Donnell et al., 2004). However, there is also evidence that depression may occur as a distinct disorder with unique risk factors in the initial months after trauma (O’Donnell et al., 2004). Given the lack of clear findings and the over-reliance on cross-sectional designs, additional studies are needed to understand how co-occurring symptoms change over time.

It is also unclear whether the relationship between PTSD and depression differs according to type of trauma. Co-occurring PTSD and depression has mostly been studied among accident survivors. As such, relatively little is known about the relationship between PTSD and depression among survivors of other trauma types, notably interpersonal trauma, such as sexual assault. Unfortunately, most studies on sexual assault survivors have also been cross-sectional and have focused on those seeking treatment for PTSD (e.g., Taft et al., 2009).

Another methodological concern is that much of the research on co-occurring PTSD and depression has evaluated comorbidity in a narrow manner, relying on categorical diagnoses only (e.g., Ikin et al., 2010, Kessler et al., 1995). Dichotomizing caseness fails to take into account clinically significant subthreshold symptoms and may underestimate the co-occurrence of symptoms and misrepresent the relationship between PTSD and depression (Backenstrass et al., 2006, Grubaugh et al., 2005). A few studies have used factor analysis to characterize the latent constructs underlying PTSD and depression symptoms. Two factor analytic studies found that PTSD loads onto the same higher-order factor as depression (Cox et al., 2002, Slade and Watson, 2006). In contrast, other factor analyses found that only some PTSD symptoms load onto a depression factor, while most PTSD and depression symptoms load onto two distinct but highly correlated factors (Gros et al., 2010). Factor analysis, however, may overlook differences in factor structure among heterogeneous groups of individuals.

Latent class analysis (LCA) is a person-centered method that accounts for heterogeneity among individuals and may therefore shed additional light on co-occurring symptom presentations after trauma. While factor analysis describes how variables group together, LCA is used to determine how individuals naturally group together into latent classes, based on shared symptom patterns (McCutcheon, 1987). LCA also utilizes the full range of symptoms instead of relying on categorical diagnoses. One type of LCA – latent profile analysis (LPA) – has the additional advantage of using continuous indicators of symptom severity, rather than dichotomous indicators reflecting the presence or absence of each symptom. Recent LCA of PTSD symptoms identified three distinct classes: no disturbance, intermediate disturbance, and pervasive disturbance, with elevated emotional numbing distinguishing the pervasive class from the intermediate class and predicting functional impairment (Breslau et al., 2005). Although LCA has been used to investigate other co-occurring symptom presentations (e.g., van Lang et al., 2006), to our knowledge it has not been used to examine co-occurring PTSD and depression symptoms among trauma survivors.

We used LPA to examine the distinctiveness of PTSD and depression after sexual assault and to identify subgroups that may be differentiated based on symptom profiles. We hypothesized that LPA would reveal distinct subgroups with different levels of co-occurring symptom severity but would not find subgroups with primarily PTSD or depression symptoms without comparable levels of the other. To assess the relationship between specific PTSD sub-clusters and depression symptoms, we analyzed PTSD symptoms based on a well-validated 4-factor model that groups symptoms into four categories: re-experiencing, avoidance, numbing, and hyperarousal (Asmundson et al., 2000, King et al., 1998).

Our second aim was to investigate whether the number or quality of latent profiles changes over time, which would suggest that the relationship between PTSD and depression symptoms is dynamic. To this end, we assessed PTSD and depression symptoms at four time-points (1, 2, 3, and 4 months post-assault) and compared LPA results from each point. Previous research on trauma survivors suggests that depression may be distinct from PTSD in the early months post-trauma, but that the distinction between these symptom presentations may fade over time (O’Donnell et al., 2004). If a similar pattern is found, a greater number of latent classes might emerge at earlier time points compared to later time points. We also examined whether individuals tend to remain in the same latent class or change classes over time. We expected to see moderate stability in class membership over time, since prior research with the study sample described below found that some individuals experience enduring PTSD symptoms while most gradually recover (Steenkamp et al., 2012).

Our third aim was to test whether different profiles of PTSD and depression symptoms prospectively predict meaningful differences in functional impairment at the final follow-up interval. If specific latent profiles of PTSD and depression symptoms 1-month after assault predict functional impairment three months later, this would have implications for identifying and intervening early with sexual assault survivors who are most likely to experience long-term functional impairment.

Section snippets

Participants and procedure

Secondary analyses were conducted on data from a study of sexual assault survivors (Steenkamp et al., 2012). Participants (N=119) were adult females recruited via Craigslist advertisements (88%) or Boston Area Rape Crisis Center advocates. Participants met inclusion criteria if they were at least 18 years of age and reported being sexually assaulted in the past month. Age ranged from 18 to 65 years (M=33.0 years; SD=10.55). Participants self-identified as Caucasian (63%), African American

Descriptive analyses

Mean DASS-D and PCL scores are presented in Table 1 (Total Sample column). As previously reported (Steenkamp et al., 2012), based on a cutoff score of 44 (Blanchard et al., 1996), probable PTSD prevalence was 78% at Month 1, 67% at Month 2, 48% at Month 3, and 41% at Month 4. At each time point, DASS-D score correlated significantly with PCL score (r=0.68−0.73, P<0.001).

Identification and description of LPA model at Month 1

Table 2 presents the fit statistics for 1- to 6-class solutions at Month 1. A 4-class solution yielded the best fit for the

Discussion

We used LPA to examine patterns in PTSD and depression symptoms across the first four months following sexual assault. Four distinct subgroups were identified at each time point, reflecting mild, low-moderate, high-moderate, and severe levels of co-occurring symptoms. Within each subgroup, the severity of PTSD and depression symptoms cohered tightly. The number of classes identified by LPA did not change over time, and at no point were there reliable subgroups containing individuals with

Role of funding source

None of the funding sources played a role in the analysis or interpretation of data; in the writing or editing of this paper; or in the decision to submit the paper for publication.

Conflict of interest

The authors have no conflicts of interests to report.

Acknowledgments

This research was partially funded by the Boston Area Rape Crisis Center and a VA Career Development Award to K. Salters-Pedneault. The authors wish to thank Anthony Rosellini for his helpful comments on an earlier version of this work.

References (38)

  • N. Breslau et al.

    Trauma and posttraumatic stress disorder in the community: the 1996 Detroit area survey of trauma

    Archives of General Psychiatry

    (1998)
  • N. Breslau et al.

    The structure of posttraumatic stress disorder: latent class analysis in 2 community samples

    Archives of General Psychiatry

    (2005)
  • C.R. Brewin et al.

    Meta-analysis of risk factors for posttraumatic stress disorder in trauma-exposed adults

    Journal of Consulting and Clinical Psychology

    (2000)
  • E. Bromet et al.

    Risk factors for DSM-III-R posttraumatic stress disorder: findings from the national comorbidity survey

    American Journal of Epidemiology

    (1998)
  • R.A. Bryant et al.

    The psychiatric sequelae of traumatic injury

    American Journal of Psychiatry

    (2010)
  • B.J. Cox et al.

    Posttraumatic stress disorder and the structure of common mental disorders

    Depression and Anxiety

    (2002)
  • T.A. deRoon-Cassini et al.

    Psychopathology and resilience following traumatic injury: a latent growth mixture model analysis

    Rehabilitation Psychology

    (2010)
  • D.F. Gros et al.

    Specificity of posttraumatic stress disorder symptoms: an investigation of comorbidity between posttraumatic stress disorder symptoms and depression in treatment-seeking veterans

    Journal of Nervous and Mental Disease

    (2010)
  • A.L. Grubaugh et al.

    Subthreshold PTSD in primary care: prevalence, psychiatric disorders, healthcare use, and functional status

    Journal of Nervous and Mental Disease

    (2005)
  • Cited by (93)

    View all citing articles on Scopus
    View full text