Identifying latent profiles of posttraumatic stress and major depression symptoms in Canadian veterans: Exploring differences across profiles in health related functioning

https://doi.org/10.1016/j.psychres.2015.03.011Get rights and content

Highlights

  • We assessed the latent groupings of veterans responding to PTSD and MDD items.

  • Resultant groupings were assessed in relation to health related functioning.

  • A 3-class LPA model was found to provide the best fit to the data.

  • Health related functioning was poorer is more symptomatic groupings.

Abstract

Posttraumatic stress disorder (PTSD) has been consistently reported as being highly comorbid with major depressive disorder (MDD) and as being associated with health related functional impairment (HRF). We used archival data from 283 previously war-zone deployed Canadian veterans. Latent profile analysis (LPA) was used to uncover patterns of PTSD and MDD comorbidity as measured via the PTSD Checklist-Military version (PCL-M) and the Patient Health Questionnaire-9 (PHQ-9). Individual membership of latent classes was used in a series of one-way ANOVAs to ascertain group differences related to HRF as measured via the Short-Form-36 Health Survey (SF-36). LPA resulted in three discrete patterns of PTSD and MDD comorbidity which were characterized by high symptoms of PTSD and MDD, moderate symptoms, and low symptoms. All ANOVAs comparing class membership on the SF-36 subscales were statistically significant demonstrating group differences across levels of HRF. The group with the highest symptoms reported the worst HRF followed by the medium and low symptom groups. These findings are clinically relevant as they demonstrate the need for continual assessment and targeted treatment of co-occurring PTSD and MDD.

Introduction

There has been a recent acknowledgment by the National Institute of Mental Health (Research Domain Criteria; RDoc) and the American Psychiatric Association (APA) that that there is an indistinct boundary between trauma or stressor-related disorders, such as Posttraumatic Stress Disorder (PTSD) and alternative disorders such as Major Depressive Disorder (MDD: APA, 2013). Indeed national epidemiological surveys estimate comorbidity rates between PTSD and MDD to range from 48% to 55% (NCS: Kessler et al., 1995; NCS-R: Kessler and Merikangas, 2004). Furthermore, both MDD and PTSD share a number of risk factors and behave similarly longitudinally (O׳Donnell et al., 2004, Norman et al., 2011). Of concern, individuals who report co-occurring PTSD and MDD are also generally those who have a more chronic profile, greater distress and impairment, and utilize mental and physical health services to a greater degree (Post et al., 2001, Ikin et al., 2010). A recent meta-analysis of 57 studies (N=6670) reported that over half of the individuals with PTSD (52%) also reported co-occurring MDD; this was more apparent in military compared to civilian samples. The current study will therefore assess the manner in which PTSD and MDD symptoms co-occur and how co-occurrence relates to health related functional impairment (HRF) in a sample of treatment seeking veterans.

A recent review of the etiology of depression comorbidity in combat-related PTSD (Stander et al., 2014) which focused on longitudinal and twin-studies, summarized three commonly proposed hypotheses which have been put forth in the extant literature. The first centers around causal hypotheses; the view that depression causes PTSD or that PTSD causes depression, the second, the common factors hypotheses; the view that both depression and PTSD result due to common risk factors, albeit the two disorders are distinct, and the third, that potential confounds may have a role in the notable comorbidity patterns; this view suggests that the two disorders are coincidentally associated. For example, the indistinct boundary between both may be attributable to poor definitions of symptomatology, thus the disorder itself (Stander et al., 2014). In conclusion, the researchers noted that more support is available for the idea that PTSD influences the development of depression compared to the reverse. In addition, it was found that common risk and vulnerabilities do indeed influence the development of PTSD and depression; yet it was acknowledged that further investigation is required. Interestingly, the authors highlighted inconsistent evidence that combat exposure may be a common risk factor for PTSD and depression. Addressing the potential confounds hypotheses it was noted that “…it is unlikely that PTSD and depression are completely indistinct diagnoses…or that they are simply sub-clusters of a larger syndrome of response to trauma…” (Stander et al., 2014; p. 94).

A number of cross-sectional studies assessing PTSD and MDD׳s comorbidity have relied on the statistical technique of Factor Analysis (Armour and Shevlin, 2010, Gros et al., 2010, Armour et al., 2011, Elhai et al., 2011a, Elhai et al., 2011b, Contractor et al., 2014). We would however argue that techniques such as Latent Class and Latent Profile Analysis (LCA and LPA, which differ in their use of categorical or continuous indictors respectively) may result in a greater degree of understanding in relation to how these two disorders co-occur. For example, LPA groups individuals into latent classes based on their similarity in responding across a number of indicators (McCutcheon, 1987). Therefore, LPA is an ideal tool to examine comorbidity given it will identify if, for example, a proportion of the sample endorses only PTSD, only MDD, or a combination of both items. Moreover, using multiple indicators allows investigators to assess how PTSD and MDD associate across all of the individual symptoms of both disorders. LPA is further advantageous in its use of continuous indicators which allow for investigations into item severity. Thus, LPA can alleviate previous methodological limitations across studies which were reliant on discrete diagnostic categories or the sole presence and absence of symptoms.

These methods have been previously utilized when examining heterogeneity in PTSD (Breslau et al., 2005, Chung and Breslau, 2008, Ayer et al., 2011, Elhai et al., 2011a, Elhai et al., 2011b) and MDD (Carragher et al., 2010; Hybels et al., 2013) symptom presentation separately. Furthermore, several studies have utilized LCA and LPA to examine how PTSD symptoms co-occur with both psychotic (Shevlin et al., 2010) and dissociative symptoms (Wolf et al., 2012a, Wolf et al., 2012b, Armour et al., 2014a, Armour et al., 2014b). To our knowledge, only a single study has conducted LPA on both PTSD and MDD symptoms concurrently (Au et al., 2013).

Au et al. (2013), in assessing PTSD and MDD data from sexual assault survivors, utilized sum scores for four well established PTSD factors (King et al., 1994 [Re-experiencing, Avoidance, Emotional Numbing, Hyperarousal]) and a measure of MDD as their LPA indicators. A 4-class model provided the best fit to the data across multiple time points. The classes highlighted that levels of PTSD and MDD mirrored each other. Most notable was that classes remained moderately stable over time and the LPA did not find a group of individuals who endorsed only PTSD or only MDD indicators. It was noted that these findings supported an earlier proposition that both disorders may represent a general distress response to traumatic experiences (O׳Donnell et al., 2004, Norman et al., 2011); with perhaps the experience of sexual assault being a common risk factor leading to the development of the two disorders (Au et al., 2013).

The current study was additionally interested in how groupings of PTSD and MDD symptoms relate to HRF. Research has consistently highlighted the relationship between trauma exposure and poor HRF (Green and Kimerling, 2003), the correspondence between PTSD symptom severity and the degree of physical health complaints (Kimerling et al., 2000), and the increased risk of morbidity in PTSD diagnosed participants, in particular veterans (Boscarino, 1997, Schnurr and Spiro, 1999, Wagner et al., 2000). Similarly, research has noted that not only do depressed patients show a greater degree of physical disability compared to non-depressed patients but that MDD has also a detrimental effect on long-term physical outcomes (Ormel et al., 1993, Beekman et al., 1995, Judd et al., 1996, Gallo et al., 1997 ). MDD may also indirectly affect HRF via an increase in health risk behaviors such as smoking, over-eating, and sedentary lifestyles (Glassman et al., 1990, Rosal et al., 2001, Goodman and Whitaker, 2002). Furthermore, comorbidity of PTSD and MDD has a greater impact on HRF (Post et al., 2001, Momartin et al., 2003) and results in long-term functional impairment (Au et al., 2013).

The first aim was to uncover underlying sub-groupings of PTSD and MDD endorsements. The second aim was to assess differential relations between groupings and reported levels of HRF. We expected to find (1) multiple PTSD and MDD groups which could be differentiated by the severity of their symptoms (Au et al., 2013) and (2) their level of HRF.

Section snippets

Participants/procedure

The current study used archival data of 424 treatment-seeking Canadian Forces (CF) members and veterans who were referred by their primary care physician (for currently serving members) or Veterans Affairs Canada case manager (for veterans) to the Parkwood Hospital Operational Stress Injury (OSI) clinic between January 2002 and May 2012.

The clinic follows a standardized intake screening protocol, which includes administering several measures including those mentioned below, to each veteran

Results

In the effective sample, the percentage of participants who met the screening criteria for probable MDD, dysthymic disorder and Depressive Disorder NOS based on PRIME MD were, 48.8 (n=138), 2.5 (n=7), and 2.5 (n=7). Further the percentage of people diagnosed with Probable PTSD based on the PTSD Checklist-Military Version (PCL-M) was 56.9 (n=161). The mean scores for the SF-36 subscales were as follows: physical functioning (M=42.17, S.D.=26.1), role limitations due to physical health (M=31.88,

Discussion

Using LPA we were able to identify three distinct classes based on how participants responded across 17 PTSD and nine MDD items in our sample of Canadian military veterans. This differs to a previous study assessing PTSD and MDD co-occurrence (Au et al., 2013) in which four latent classes were deemed optimal. Notably, both the current study and Au et al.׳s study reported classes with increasing symptom severity in the absence of qualitative differences in item endorsement. There were however

Author disclosures

This manuscript received no funding. We have no acknowledgement to makes. Nor do we have any conflicts of interest to report.

References (67)

  • S.D. McDonald et al.

    The diagnostic accuracy of the PTSD Checklist: a critical review

    Clinical Psychology Review

    (2010)
  • S. Momartin et al.

    Dimensions of trauma assocaiated with posttraumatic stress disorder PTSD caseness, severity and functional impairment: a study of Bosnian refugees resettled in Australia

    Social Science and Medicine

    (2003)
  • M.C. Rosal et al.

    Behavioral risk factors among members of a health maintenance organization

    Preventative Medicie

    (2001)
  • C. Yang

    Evaluating latent class analysis models in qualitative phenotype identification

    Computational Statistics and Data Analysis

    (2006)
  • American Psychiatric Association

    Diagnostic and Statistical Manual Of Mental Disorders

    (2013)
  • C. Armour et al.

    The co-occurrence of PTSD and dissociation: differentiating severe PTSD from dissociative-PTSD

    Social Psychiatry and Psychiatric Epidemiology

    (2014)
  • C. Armour et al.

    Testing the robustness of the dysphoria factor of the Simms et al. (2002) model of PTSD

    Psychological Trauma: Theory, Research Practice and Policy

    (2011)
  • C. Armour et al.

    Testing the dimensionality of PTSD and the specificity of the dysphoria factor

    Journal of Loss and Trauma

    (2010)
  • C.G. Beevers et al.

    Association of predeployment gaze bias for emotion stimuli with later symptoms of PTSD and depression in soldiers deployed in Iraq

    American Journal of Psychiatry

    (2011)
  • J.A. Boscarino

    Diseases among men 20 years after exposure to severe stress: implications for clinical research and medical care

    Psychosomatic Medicine

    (1997)
  • N. Breslau et al.

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

    Archives of General Psychiatry

    (2005)
  • C.R. Brewin et al.

    Recognition and treatment of psychological disorders during military service in the UK armed forces: a study of war pensioners

    Social Psychiatry and Psychiatric Epidemiology

    (2012)
  • H. Chung et al.

    The latent structure of post-traumatic stress disorder: tests of invariance by gender and trauma type

    Psychological Medicine

    (2008)
  • J. Cohen

    Statistical Power Analysis for the Behavioral Sciences

    (1988)
  • C. DiStefano et al.

    Investigating subtypes of child development. A comparison of cluster analysis and latent class cluster analysis in typology creation

    Educational and Psychological Measurement

    (2006)
  • J.D. Elhai et al.

    Heterogeneity in clinical presentations of posttraumatic stress disorder among medical patients: testing factor structure variation using factor mixture modeling

    Journal of Traumatic Stress

    (2011)
  • D.J. Erickson et al.

    Posttraumatic stress disorder and depression symptomatology in a sample of Gulf War veterans: a prospective analysis

    Journal of Consulting and Clinical Psychology

    (2001)
  • J.J. Gallo et al.

    Depression without sadness: functional outcomes of nondysphoric depression in later life

    Journal of American Geriatric Society

    (1997)
  • M.W. Gilbertson et al.

    Smaller hippocampal volume predicts pathologic vulnerability to psychological trauma

    Nature Neuroscience

    (2002)
  • A.H. Glassman et al.

    Smoking, smoking cessation, and major depression

    JAMA

    (1990)
  • E. Goodman et al.

    A prospective study of the role of depression in the development and persistence of adolescent obesity

    Pediatrics

    (2002)
  • B.L. Green et al.

    Trauma, posttraumatic stress disorder and health status

  • D.F. Gros et al.

    Comparing the symptoms of posttraumatic stress disorder with the distress and fear disorders

    Journal of Nervous Mental Disease

    (2012)
  • Cited by (49)

    • Severity profiles of posttraumatic stress, depression, anxiety, and somatization symptoms in treatment seeking traumatized refugees

      2020, Journal of Affective Disorders
      Citation Excerpt :

      In particular refugees face significant post-migration, displacement-related stressors that have a significant negative influence on mental health problems (Carswell et al., 2011; Miller and Rasmussen, 2017; Minihan et al., 2018). Because exposure to past PTEs brings about increased and enhanced sensitivity to post-trauma stressors (Smid et al., 2013), enduring trauma-related mental health problems combined with lasting displacement-related stressors will cause a gradual decline of adaptive psychosocial and health related functioning and will lead to a broad mix of general psychiatric symptoms and disorders instead of clearly distinguishable diagnostic classifications (Au et al., 2013; Armour et al., 2015; Jongedijk et al., 2019). Although we measured a broad spectrum of comorbid psychopathology next to PTSD, some potentially important symptom dimensions were not taken into account, including affective dysregulation, dissociation, prolonged grief, and personality dimensions.

    View all citing articles on Scopus
    View full text