Elsevier

Journal of Affective Disorders

Volume 169, 1 December 2014, Pages 40-46
Journal of Affective Disorders

Research report
PTSD prevalence and symptom structure of DSM-5 criteria in adolescents and young adults surviving the 2011 shooting in Norway

https://doi.org/10.1016/j.jad.2014.06.055Get rights and content

Abstract

Background

Diagnostic criteria for Posttraumatic Stress Disorder (PTSD) have been revised for DSM-5. Two key changes include alteration of the clustering of PTSD symptoms and new PTSD symptom criteria related to negative alterations in cognition and mood. In this study, we empirically investigated these changes.

Methods

We interviewed 325 adolescents and young adults who survived the 2011 youth camp shooting at Utøya Island, Norway. The UCLA PTSD Reaction Index for DSM-IV was used to assess symptoms of PTSD. In addition, 11 questions were added to assess the four new symptom criteria within the new DSM-5 symptom categories.

Results

PTSD prevalence did not differ significantly whether DSM-IV (11.1%) or DSM-5 (11.7%) criteria were used and the Cohen׳s Kappa for consistency between the diagnoses was 0.061. Confirmatory factor analyses showed that the four-factor structure of the DSM-5 fit the data adequately according to the conceptual model outlined.

Limitations

The homogeneity of this sample of highly exposed subjects may preclude generalization to less severely exposed groups. Also, we did not assess criterion G in regard to symptoms causing clinically significant distress and functional impairment.

Conclusion

The prevalence of PTSD was quite similar regardless of diagnostic system. The relatively low concordance between the diagnoses has implications for eligibility for a diagnosis of PTSD.

Introduction

The diagnosis of posttraumatic stress disorder (PTSD) has been a subject of much debate since it was introduced in the Diagnostic and Statistical Manual (DSM-III, American Psychiatric Association, 1980), and has gone through several modifications since that time. Since the introduction of PTSD in the fourth edition of the Diagnostic and Statistical Manual for Mental Disorders (APA, 1987), the diagnosis has been characterized by three symptom clusters: (B) Reexperiencing; (C) Avoidance and Emotional Constriction; and (D) Arousal. The fifth edition was published in May, 2013. Changes from the previous version in the symptom criteria for PTSD include (1) that symptoms of avoidance were separated out as its own category; (2) a new category was introduced that included negative alterations in cognition and mood; and (3) new symptoms were added to include marked alterations in arousal and reactivity, specifically angry outbursts and reckless or self-destructive behavior. These changes may have implications for diagnostic prevalence and for the underlying structure of PTSD. In this paper, we explore the changes in the diagnostic criteria, as well as the underlying symptom structure among 325 adolescents and young adults surviving a terrorist attack targeting a youth camp on the Norwegian island Utøya in 2011.

The most notable change from DSM-IV to DSM-5 is the inclusion of a new symptom category covering a set of negative alterations in cognition and mood. This category includes seven criteria, three retained from the DSM-IV while separated from avoidance symptoms, and four new added criteria. These encompass negative expectations about oneself, others and the world expanded well beyond the DSM-IV symptom of foreshortened future, distorted blame of self or others, pervasive negative emotional state as well as the cognitive/dissociative symptom of distorted memory about important aspects of the event. The inclusion of symptoms that previously have not been considered part of a PTSD cluster has generated debate. First, Brewin et al. (2009) have expressed concern over the inclusions of too wide a range of dysphoric states that would decrease diagnostic specificity. Second, there is debate about whether these new sets of criteria are likely to change prevalence, either by including a number of individuals previously not eligible for this diagnosis (Calhoun et al., 2012), and/or exclude other individuals who were previously eligible for a PTSD diagnosis.

A few studies have so far evaluated the possible change in PTSD prevalence with the altered symptom criteria. In the initial field trial of the proposed DSM-5 criteria through an internet-based survey, Kilpatrick (2013) found a small decrease in the prevalence of PTSD using DSM-5 criteria that was primarily due to a tightening of the A1 gateway (trauma exposure), especially in regard to restricting “learning about” to only violent and accidental death. Similar results emerged in studies evaluating the changed PTSD symptoms in a large population and a sample of military veterans (Miller et al., 2013) as well as in an adult primary care sample (Contractor et al., 2014). In a non-clinical sample of college students Elhai et al. (2012) found that overall, the DSM-5 symptoms provided a somewhat higher PTSD prevalence than did the DSM-IV conceptualization (Elhai et al., 2012). For both DSM-IV and DSM-5 the prevalence decreased considerably when requirement of impairment was included. Overall, the few studies conducted so far have suggested that the transition to a new diagnostic system does not influence the prevalence notably. Including new symptoms into a diagnosis may not only influence prevalence however, but may also alter diagnostic eligibility i.e. whether the same individuals will be diagnosed using either of the diagnostic systems. Only two studies have so far evaluated the overlap between the two diagnostic systems, both finding a relatively high concordance (Calhoun et al., 2012, Carmassi et al., 2013).

Examining the underlying symptom structure may help to further elucidate the psychopathology that underlies PTSD. There is considerable evidence that the tree-factor structure in DSM-IV is not as well-supported as two four-factor solutions that have proved to better represent the latent structure of PTSD symptoms using the DSM-IV (King et al., 1998, Simms et al., 2002). King et al. (1998) developed the four-factor emotional numbing model, separating DSM-IV׳s PTSD avoidance (C1–C2) and numbing (C3–C7) factors. Simms et al. (2002), on the other hand, noted that several of the PTSD symptoms reflect a more general distress component. Based on this, they developed the four-factor dysphoria model, in which three of the hyperarousal symptoms (sleep, irritability and concentration problems) are placed into a separate dysphoria factor while the remaining hyperarousal symptoms are retained as a hyperarousal factor. Both models have received substantial empirical support in various samples of trauma-exposed individuals (Goozeit and Markon, 2011, Yufik and Simms, 2010), although none of them has proved consistently better than the other (Elhai and Palmieri, 2011). A third factor model that has received considerable support is the five-factor dysphoric arousal model (Elhai et al., 2011), in which the arousal factor is separated into one factor reflecting a dysphoric arousal (sleep, irritability and concentration) and a second factor reflecting anxious arousal (hypervigilance and exaggerated startle response). The dysphoric arousal model has generally proved superior to other models (Armour et al., 2012, Elhai et al., 2011). The development and evaluation of these models have been instrumental in the revision of the PTSD diagnosis in the DSM-5 and in development of the new four-factor conceptual model for PTSD. The DSM-5 revision bears similarity to the DSM-IV numbing model of King et al. (1998), in that the effortful avoidance symptoms (C1 and C2) are separated from the other symptoms reflecting “numbing of general responsiveness” (Friedman et al., 2011). With emotional numbing now included in a larger D Category, along with other new diagnostic criteria, this is an opportune time to evaluate the factor structure of PTSD in DSM-5. The revisions incorporated in DSM-5 were designed to take account of scientific evidence and to be responsive to a growing outcome of intervention studies of PTSD with protocols that successfully address negative appraisals, expectations and emotions (Friedman et al., 2011). Importantly, examining the factor structure of PTSD in DSM-5 may help to further elucidate the psychopathology and core elements that underlie PTSD, improve clinical assessment, identification of intervention foci, and delineate symptoms that most relate to distress and impairment over time. A few studies have so far evaluated the latent structure of DSM-5 conceptualization. Miller et al. (2013) compared four competing models in a sample of veterans, finding that the DSM-5 model had a better fit than three alternative models. Two studies have found support for the DSM-5 model (Biehn et al., 2013, Contractor et al., 2014), while Miller et al., (2013) found a slight support for a modified dysphoria model. In a recent study of Chinese earthquake survivors the DSM-5 criteria were evaluated through a comparison of six alternative factor models, including the DSM-5 conceptual four-factor model, a dysphoria model and a dysphoric arousal model, in addition to three alternative solutions (Liu et al., 2014). In that sample, all solutions provided a good fit with the data, with a slight support for a six-factor revision of the DSM-5 dysphoric arousal model splitting criterion D (negative alterations of cognition and mood) into two separate factors: D1–D4 into a separate factor of negative affect and D5–D7 into an anhedonia factor. Overall, the four-factor solution for DSM-5 has shown promising results, but there is need for accumulating more evidence from diverse samples to evaluate the underlying symptom structure of the DSM-5 conceptualization. In particular, there is a need for evaluating whether this conceptualization reflects the symptom structure in adolescents as such knowledge is still lacking. We were interested in evaluating this new PTSD conceptualization compared to the previously well-supported dysphoria and dysphoric arousal models. In light of recent research suggesting negative affect and anhedonia may reflect to distinct phenomena (Liu et al., 2014); we also evaluated a five-factor model separating these two constructs.

The transition to a new diagnostic system has potential consequences for deriving prevalence rates of PTSD, as well as the latent structure of PTSD symptoms. To date, only a few studies have explored these changes (Calhoun et al., 2012, Carmassi et al., 2013, Kilpatrick et al., 2013, Miller et al., 2012), and in particular among youth (Elhai et al., 2012). In the present study, we evaluated the new DSM-5 diagnostic criteria for PTSD by assessing the prevalence rate of PTSD and evaluating the factor structure of the new criteria among a highly-exposed sample of adolescents and young adults.

Section snippets

Participants and procedures

The police registered 495 survivors of the terrorist attack on Utøya Island that occurred on July 22, 2011. Three months after the terrorist attack, the 490 survivors who were at least 13 years of age were sent postal invitations to participate in the present study and were subsequently contacted by telephone. One-hundred and sixty-five survivors could not be reached by telephone or declined to participate. As a result, 325 (66.3%) survivors were interviewed face-to-face, most of them in their

Sample characteristics

Of the 325 participants in this sample, just over half were boys (N=172, 52.9%), and 92% of the participants were less than 25 years of age (M=19.3, SD=4.6). The majority was ethnically Norwegian, and 11% had a minority background. Most of them were students at least half-time (N=235, 81.1%), and a minority were employed in full- or part-time work (N=87, 28.4). The number of potentially traumatizing elements participants were exposed to during the attack was on average high; no participants

Discussion

This study evaluated DSM-5 diagnostic criteria for PTSD in a sample of highly traumatized youth. Using a conservative threshold for symptom presence, PTSD prevalence rates were quite similar for either the DSM-IV or DSM-5 algorithms. (11.1% vs. 11.7%, respectively). Future studies should investigate whether, for selected PTSD symptoms, the threshold for endorsement might be at a lower frequency of occurrence because the symptoms may be associated with serious consequences, for example,

Role of funding source

The study was funded by the Norwegian Directorate of Health.

Conflict of interest

All authors declare that they have no actual or potential conflicts of interest.

Acknowledgments

We thank the survivors of the terror attack at Utøya Island who took part in this study, and acknowledge the effort of professionals all over Norway who conducted the interviews. We would also like to thank Dr. Dean Kilpatrick for valuable advice and suggestions on a previous draft of this paper. The study was funded by the Norwegian Directorate of Health.

References (34)

  • C.R. Brewin et al.

    Reformulating PTSD for DSM-V: Life after Criterion A

    J. Trauma. Stress

    (2009)
  • P.S. Calhoun et al.

    The effect of draft DSM-V criteria on posttraumatic stress disorder prevalence

    Depress. Anxiety

    (2012)
  • M.J. Friedman et al.

    Considering PTSD for DSM-5

    Depress. Anxiety

    (2011)
  • J. Goozeit et al.

    Factors of PTSD: differential specificity and external correlates

    Clin. Psychol. Rev.

    (2011)
  • L.-T. Hu et al.

    Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives

    Struct. Equ. Model.

    (1999)
  • D.G. Kilpatrick

    The DSM-5 got it right: comment on Friedman (2013)

    J. Trauma. Stress

    (2013)
  • D.G. Kilpatrick et al.

    National estimates of exposure to traumatic events and PTSD prevalence using DSM-IV and DSM-5 criteria

    J. Trauma. Stress

    (2013)
  • Cited by (53)

    • Mental health after mass shootings

      2023, Encyclopedia of Mental Health, Third Edition: Volume 1-3
    • Elucidating posttraumatic stress symptom dimensions and health correlates among postpartum women

      2021, Journal of Affective Disorders
      Citation Excerpt :

      In 2011, a five-factor dysphoric arousal model categorized the D1—D3 symptoms into a separate latent factor representing dysphoric arousal and shifted other, fear-based hyperarousal symptoms into an anxious arousal factor, resulting in five dimensions: re-experiencing, avoidance, numbing, anxious arousal, and dysphoric arousal (Elhai et al., 2011). A recent review of over 100 studies in a variety of samples found that this five-factor model demonstrates superior fit over both four-factor models (Armour et al., 2016), even among studies utilizing revised DSM-5 PTSD criteria (Gentes et al., 2014; Hafstad et al., 2014; Tsai et al., 2015). While a fair amount of research has focused on PTSD symptom dimensions among veterans (Harpaz-Rotem et al., 2014; Pietrzak et al. 2010; Tsai et al., 2015) and civilian populations such as World Trade Center responders (Pietrzak et al., 2014) and cardiac arrest patients (Presciutti et al., 2019), minimal research has examined the dimensionality of PTSD in postpartum women.

    • Exploring mood symptoms overlap in PTSD diagnosis: ICD-11 and DSM-5 criteria compared in a sample of subjects with Bipolar Disorder

      2020, Journal of Affective Disorders
      Citation Excerpt :

      All the excluded not-specific symptoms were found, in fact, significantly more reported in subjects with ICD-11 PTSD than in those without. These results corroborate previous findings from studies on DSM-5 PTSD on the pivotal role of these symptoms, regardless their overlapping nature with mood disorders (Friedman et al., 2011; Carmassi et al., 2013; Dell'Osso et al., 2013; Hafstad et al., 2014; Afzali et al., 2017; Durham et al., 2018; Barbano et al., 2019). A recent network analysis study showed how putatively non-specific symptoms (e.g. persistent negative emotional state and inability to experience negative emotions) are indeed the predominant features in PTSD and represent a cause of clinical distress, whether they are listed among diagnostic criteria or not (Mitchell et al., 2017).

    • A network analysis of DSM-5 posttraumatic stress disorder and functional impairment in UK treatment-seeking veterans

      2018, Journal of Anxiety Disorders
      Citation Excerpt :

      Supporting the results of Armour, Fried, Deserno et al. (2017) and Armour, Fried, Olff et al. (2017), we found that symptom D1 (memory impairments) had the smallest number of connections, which were also relatively weak. This finding is consistent with the factor analytic literature, in which the memory impairments symptom has been consistently found to load weakly on the emotional numbing (DSM-IV studies) or the negative alterations in cognitions and mood (DSM-5 studies) symptom cluster (e.g., Hafstad, Dyb, Jensen, Steinberg, & Pynoos, 2014; Price and van Stolk-Cooke, 2015; Seligowski & Orcutt, 2016; Yufik and Simms, 2010). The low connectivity of the memory impairments symptom was further supported in the centrality analyses, where this symptom was significantly less central than all other PTSD symptoms (except for symptom E2–reckless or self-destructive behaviour).

    View all citing articles on Scopus
    View full text