Assessing possible DSM-5 ASD subtypes in a sample of victims meeting caseness for DSM-5 ASD based on self-report following multiple forms of traumatic exposure

Highlights

• A 5-class LPA model was found to provide the best fit to the data.

• Results suggest that ASD may include a dissociative avoidant subtype.

• The existence of ASD subtypes may call for individualized tailored treatment.

Abstract

Acute stress disorder (ASD) was introduced into the DSM-IV to recognize early traumatic responses and as a precursor of PTSD. Although the diagnostic criteria for ASD were altered and structured more similarly to the PTSD definition in DSM-5, only the PTSD diagnosis includes a dissociative subtype. Emerging research has indicated that there also appears to be a highly symptomatic subtype for ASD. However, the specific nature of the subtype is currently unclear. The present study investigates the possible presence of ASD subtypes in a mixed sample of victims meeting caseness for DSM-5 ASD based on self-report following four different types of traumatic exposure (N = 472). The results of latent profile analysis revealed a 5-class solution. The highly symptomatic class was marked by high endorsement on avoidance and dissociation compared to the other classes. Findings are discussed in regard to its clinical implications including the implications for the pending the ICD-11 and the recently released DSM-5.

Introduction

The acute stress disorder (ASD) diagnosis was introduced to the diagnostic and statistical manual of mental disorders 4th edition (DSM-IV) two decades ago, and ASD has subsequently been revised in the DSM-5 (DSM-IV; American Psychiatric Association, 1994, American Psychiatric Association, 2013). The ASD diagnosis was originally introduced to recognize acute posttraumatic stress symptoms and as a way of identifying victims at risk of developing posttraumatic stress disorder (PTSD; American Psychiatric Association, 1994). Notably, the idea that ASD is a precursor to PTSD, has been abandoned in the DSM-5 based on research which has repeatedly shown mixed results in relation to the capacity of the ASD diagnosis to predict the PTSD diagnosis (Bryant, 2011). One of the curious developments in DSM-5 was that ASD lost the emphasis on dissociative symptoms that initially distinguished it from PTSD in DSM-IV, and at the same time DSM-5 introduced a dissociative subtype of PTSD. In both the DSM-IV and the DSM-5, the diagnostic criteria of ASD and PTSD describe symptoms of intrusion, avoidance, and arousal. In the DSM-IV the main difference between the two diagnoses (besides the duration criterion) was that the DSM-IV ASD diagnosis required the presence of dissociative symptoms, whereas the DSM-IV PTSD diagnosis did not. The DSM-IV defines peritraumatic dissociation as a subjective feeling of emotional numbness, detachment from others, reduced responsiveness to one's surroundings, depersonalization, and derealization during the traumatic exposure (American Psychiatric Association, 1994). Dissociation has been associated with both acute and long-term posttraumatic stress symptoms across numerous forms of traumatic exposure (cf. Breh and Seidler, 2007, Cardeña and Carlson, 2011, Ozer et al., 2003). The importance of dissociation in relation to ASD has been reduced in the DSM-5 compared to the DSM-IV, since the DSM-5 ASD diagnosis no longer requires the presence of a specific number of dissociative symptoms, rather just the presence of 9 out of a total of 14 symptoms (i.e. those belonging to the five categories of intrusion, arousal, negative mood, avoidance, and dissociation). This change was made on the basis of increasing evidence that peritraumatic dissociation does not adequately predict PTSD with sufficient sensitivity and specificity (Bryant, 2011). The first formal recognition of dissociative presentations in longer-term traumatic responding occurred in DSM-5 because of initial evidence that PTSD patients with dissociative tendencies had distinctive neural and behavioral characteristics (Felmingham et al., 2008, Lanius et al., 2010).

Recently, several studies using latent profile analysis (LPA) have identified a number of PTSD subtypes including the dissociative subtype (cf. Armour et al., 2014a, Armour et al., 2014b, Wolf et al., 2012a, Wolf et al., 2012b). As argued by Armour and Hansen (2015), due to the close similarities in the diagnostic criteria of both disorders, severity and latent structure (cf. Classen et al., 1998, Hansen et al., 2012, Hansen and Elklit, 2013, Wang et al., 2012) as well as two reviews of ASD (Cardeña and Carlson, 2011, Isserlin et al., 2008) and the proposed ICD-11 recommendation for ASD symptoms (WHO, 2014), it is possible that a dissociative subtype may also be apparent for ASD. Furthermore, to the best of our knowledge only two studies have investigated the latent profile of ASD (Armour and Hansen, 2015, Shevlin et al., 2014). These two studies, combined with a limited body of research, suggest that alternative ASD subtypes are also likely to exist.

Using LPA on the mean scores of the four ASD DSM-IV symptom clusters, Shevlin et al. (2014) found that the latent profiles of ASD in Danish victims of rape (N = 471) differ both quantitatively and qualitatively. The results of the LPA revealed four classes: high mean scores on all four ASD symptom clusters, low mean scores on all four DSM-IV ASD symptom clusters, and two intermediate classes. The two intermediate classes were similar in the mean number of dissociation and re-experiencing symptoms, but one intermediate class had a higher number of arousal and lower number of avoidance symptoms compared to the other intermediate class. Armour and Hansen (2015) used LPA on the 19 individual symptoms of DSM-5 ASD and identified four different classes in Danish victims of bank robbery (N = 450): a highly symptomatic class, a class with moderate intrusion and high endorsement on the remaining items, a class characterized by low scores on most items except moderate to low on arousal, and a low symptomatic class. Although, the studies are not directly comparable due to different analytical approaches, combined the two studies suggest that ASD is not simply experienced by quantitative differences in severity, but instead qualitatively different profiles may also to exist. These profiles may also differ following different forms of traumatic exposure. Both studies were based on Danish victims of interpersonal violence with ASD prevalence rates of 68.8% according to the DSM-IV in rape victims and 10% according to the DSM-5 in bank robbery victims. As argued by Hansen et al. (2012) it is possible that the latent structure of ASD may differ in clinical samples compared to mixed samples. Thus, it is important to investigate the latent profile of ASD symptoms in clinical populations. At the same time, it is also possible that different subtypes of ASD exist across traumatic exposures and thus it is important to investigate the latent structure of ASD in a heterogeneous trauma sample. The existence of ASD subtypes may help to explain the mixed results regarding the capacity of ASD to predict PTSD across traumatic exposure and the confirmatory factor analytic studies failing to support the DSM-5 one-factor structure of ASD symptoms. Indeed, the PTSD latent factor structure is found to differ in veterans with and without PTSD (Biehn, Elhai, Fine, Seligman, & Richardson, 2012). Thus, the present study investigated the latent profile of ASD symptoms in a sample of victims of different traumatic exposure meeting caseness for DSM-5 ASD based on self-report. Based on the limited research we hypothesized that different ASD subtypes would exist which are both quantitatively and qualitatively different. More specifically, we expected to find a highly symptomatic class and intermediate ASD severity subgroups. We did not expect to find a low symptomatic group as we were assessing ASD subtypes in a sample meeting caseness for DSM-5 ASD based on self-report and we were uncertain about the specific number of intermediate classes. Although, previous research has reached different conclusions in regard to ASD subtypes (i.e. dissociative or intrusive), we expected that we may uncover a dissociative ASD subtype given the highly symptomatic nature of the participants in the current study.