Comparison of DSM-IV and proposed ICD-11 formulations of PTSD among civilian survivors of war and war veterans.

The World Health Organization recently proposed a reformulation of posttraumatic stress disorder (PTSD) for the 11th edition of the International Classification of Diseases (ICD‐11), employing only 6 symptoms. The aim of this study was to investigate the impact of this reformulation of PTSD as compared to criteria according to Diagnostic and Statistical Manual of Mental Disorders (4th ed., DSM‐IV; American Psychiatric Association, 1994) on the prevalence of current PTSD as well as comorbid major depressive episode and anxiety disorders other than PTSD. Study 1 involved previously collected interviews with 560 Kosovar civilian war survivors; Study 2 employed a previously collected sample of 142 British war veterans. Results revealed no change in the diagnostic status under the criteria proposed for ICD‐11 in 87.5% of civilian war survivors and 91.5% of war veterans. Participants who only met the newly proposed criteria showed lower rates of comorbid major depressive episode than participants who only met DSM‐IV criteria (13.6% vs. 43.8% respectively). Rates of comorbid anxiety disorders did not significantly differ between participants who lost or gained a PTSD diagnosis under the proposed criteria.

Resumen: La Organización Mundial de la Salud propuso una reformulación del Trastorno por Estrés Post Traumático (TEPT) para la decimoprimera edición de la Clasificación Internacional de Enfermedades (CIE‐11), empleando sólo 6 síntomas. El objetivo de este estudio fue investigar el impacto de esta reformulación del TEPT en comparación con los criterios actuales del TEPT en el DSM‐IV, así como también la comorbilidad con episodio depresivo mayor y otros trastornos ansiosos. El estudio 1 involucró la recolección de entrevistas realizadas previamente a 560 civiles Kosovianos sobrevivientes de la guerra; el estudio 2 empleó una muestra recolectada previamente de 142 veteranos de guerra británicos. El resultado reveló que no había cambio en el estatus diagnóstico bajo el criterio propuesto por la CIE‐11 en un 87,5% de los civiles sobrevivientes a la guerra y en un 91,5% de los veteranos de guerra. Los participantes que sólo cumplían con los criterios recientemente propuestos mostraron menor tasa de comorbilidad con episodio depresivo mayor que los participantes que sólo cumplían con los criterios del DSM‐IV (13.6% vs. 43.8% respectivamente). Las tasas de comorbilidad con trastornos ansiosos no difirieron significativamente entre los participantes que perdieron o ganaron el diagnóstico de TEPT bajo los nuevos criterios propuestos.

標題: DSM‐IV中和在ICD‐11中提出改寫的創傷後壓力症對比 撮要: 世界衛生組織近日就國際疾病傷害及死因分類標準第十一版(ICD‐11)提出對創傷後壓力症(PTSD)作出改寫，只採用6項症狀。本研究目的旨在調查是次改寫PTSD的影響，與⟪精神疾病診斷與統計手冊第四版⟫(DSM‐IV)就目前PTSD以及共病嚴重抑鬱節段和其他焦慮症的普遍程度準則作對比。研究一採用先前所收集科索沃戰爭560名生還者的面談；研究二採用了先前收集142名英國退役軍人的樣本。結果顯示，根據ICD‐11提出的準則，診斷狀態在百分之87.5的戰爭生還者和百分之91.5的退役軍人中沒有改變。只符合新提出準則的參與者，相比只符合DSM‐IV準則的參與者，反映出較低的共病嚴重抑鬱節段(分別為百分之13.6 比百分之43.8)。在新提出的準則中，PTSD診斷或增或減的參與者，其共病焦慮症比率則沒有明顯差別。 關鍵詞：PTSD, DSM‐5, ICD‐11, 戰爭, 創傷事件, 共病 标题: DSM‐IV中和在ICD‐11中提出改写的创伤后压力症对比 撮要: 世界卫生组织近日就国际疾病伤害及死因分类标准第十一版(ICD‐11)提出对创伤后压力症(PTSD)作出改写，只采用6项症状。本研究目的旨在调查是次改写PTSD的影响，与⟪精神疾病诊断与统计手册第四版⟫(DSM‐IV)就目前PTSD以及共病严重抑郁节段和其他焦虑症的普遍程度准则作对比。研究一采用先前所收集科索沃战争560名生还者的面谈；研究二采用了先前收集142名英国退役军人的样本。结果显示，根据ICD‐11提出的准则，诊断状态在百分之87.5的战争生还者和百分之91.5的退役军人中没有改变。只符合新提出准则的参与者，相比只符合DSM‐IV准则的参与者，反映出较低的共病严重抑郁节段(分别为百分之13.6 比百分之43.8)。在新提出的准则中，PTSD诊断或增或减的参与者，其共病焦虑症比率则没有明显差别。 关键词：PTSD, DSM‐5, ICD‐11, 战争, 创伤事件, 共病

Posttraumatic stress disorder (PTSD) was first defined in the third edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM‐III; American Psychiatric Association [APA], [1] ) and later revised in its fourth edition (DSM‐IV; APA, [2] ). According to the DSM‐IV, the diagnostic criteria for PTSD require the onset of characteristic symptoms following exposure to an extreme event and a reaction to that event that involves fear, helplessness, or horror (Criterion A1 and A2). Posttraumatic symptoms must be present for more than 1 month and include at least one reexperiencing symptom (Criterion B), three avoidance/numbing symptoms (Criterion C), and two hyperarousal symptoms (Criterion D) out of a total of 17 possible symptoms. The diagnosis of PTSD has greatly influenced research and practice, but has been subject to various criticisms (Rosen & Lilienfeld, [24] ; Spitzer, First, & Wakefield, [26] ). In this article, we present data on an alternative approach proposed for the World Health Organization's (WHO) 11th edition of the International Classification of Diseases (ICD‐11), in which the number of qualifying PTSD symptoms is reduced from 17 to 6 (Maercker et al., [16] ).

The DSM‐IV PTSD diagnosis has been criticized on at least four fronts. First, there is debate on the utility and demarcation of the stressor criterion (Criterion A) and the overlap among separate PTSD symptom criteria (Yufik & Simms, [29] ). Second, the disorder has been criticized for excessive complexity, resulting in thousands of different combinations of symptoms all leading to the diagnosis (Brewin, Lanius, Novac, Schnyder, & Galea, [8] ; Spitzer et al., [26] ). Third, criticisms have focused on symptom overlap and comorbidity with other mental disorders (Rosen & Lilienfeld, [24] ; Spitzer et al., [26] ). Fourth, concerns have been raised about the broader societal consequences of the PTSD diagnosis, such as the risk of pathologizing psychological reactions to normal distress (McHugh & Treisman, [17] ). The DSM‐5 (APA, [3] ) addresses a number of these problems and following an extensive review of the evidence has modified Criterion A and the constituent symptom clusters of PTSD, further improving symptom description and adding three new symptoms corresponding to common features of the disorder (Friedman, Resick, Bryant, & Brewin, [12] ).

The WHO's ICD first introduced the criteria for PTSD in its 10th revision (ICD‐10; World Health Organization, [28] ). The ICD‐10 also included exposure to a stressor and symptoms from each of three symptom clusters and the onset of symptoms had to be within 6 months. Yet, in ICD‐10, there was no subjective stressor criterion and there was a greater emphasis on reexperiencing and less on emotional numbing. To maximize clinical utility, proposals for an upcoming revision to ICD‐10 (ICD‐11) have included a reduction and simplification in the symptoms required for a PTSD diagnosis (Maercker et al., [16] ). An important proposed innovation in ICD‐11 involves specifying core elements rather than typical features of PTSD. Core elements are those that on empirical or theoretical grounds most clearly distinguish PTSD from other disorders. The elements had additionally to be endorsed by clinicians working in a variety of cultural settings around the world. Following suggestions by Brewin et al. ([8] ), the first core element consists of reexperiencing the traumatic event(s) in the present, as evidenced by either flashbacks or nightmares, accompanied by fear or horror (see Table [NaN] ). Consistent with DSM‐5, flashbacks are defined as intrusive waking memories in which reexperiencing in the present can vary from a transient sensation to a complete disconnection from the current environment. The second core element is avoidance of these intrusions, as evidenced by marked internal avoidance of thoughts and memories, or external avoidance of activities or situations reminiscent of the traumatic event(s). The third core element is an excessive sense of current threat, as evidenced either by hypervigilance or by exaggerated startle, two arousal symptoms that tend to cluster together (Yufik & Simms, [29] ). PTSD is defined in terms of the presence of at least one of the two symptoms from each of these three core elements, in addition to impairment in functioning. The effect is to require the presence of at least 50% (3/6) of the specified core symptoms in place of at least 35% (6/17) of the typical features described in DSM‐IV.

Proposed Diagnostic Criteria for PTSD in ICD‐11

• 2 Note
• 3 PTSD = posttraumatic stress disorder; ICD‐11 = International Classification of Diseases (11th ed.). Adapted from “Reformulating PTSD for DSM‐V: Life after Criterion A” by C. R. Brewin, R. A. Lanius, A. Novac, U. Schnyder, & S. Galea, 2009, Journal of Traumatic Stress, 22, 366–373. Copyright 2009 by the International Society for Traumatic Stress Studies.

The implications of the proposed ICD‐11 criteria on the prevalence of PTSD and on its comorbidity with other disorders are unknown. To our knowledge, only one study has evaluated the impact of these criteria on the prevalence and comorbidity pattern of PTSD (van Emmerik & Kamphuis, [27] ). In a sample of 170 treatment‐seeking survivors of civilian traumas, no significant change in PTSD prevalence was found. This study suggested that concerns about increased prevalence are not warranted and that the high comorbidity of PTSD with major depressive episode (MDE) in particular might not be explicable by symptom overlap alone.

This investigation aimed to compare the prevalence of PTSD under the DSM‐IV and proposed ICD‐11 criteria in nontreatment‐seeking samples. This comparison is possible because the newly proposed ICD‐11 criteria for PTSD constitute a subset of the PTSD symptoms specified in the DSM‐IV criteria (specifically, symptoms B2, B3, C1, C2, D4, and D5). Accordingly, instruments capturing the DSM‐IV criteria for PTSD can be used to evaluate the proposed ICD‐11 criteria.

Study 1 utilized data from two surveys conducted in Kosovo in 2009 (Morina, von Lersner, & Prigerson, [22] ; Morina & Emmelkamp, [20] ), 10 years after the war. The survey was conducted in the municipalities of Gllogovc, Kline, Skenderaj, and Vushtrri. The interviews were conducted by five female psychologists who were experienced in conducting clinical interviews for a prior project (Priebe et al., [23] ). In the prior project, interrater agreement among interviewers was assessed for the MINI International Neuropsychiatric Interview in two mock interviews. Among 251 items, the mean agreement rate (i.e., all interviewers gave the same answer for each item) across two sessions was 90.2%.

In accordance with the original aim of the study to investigate the impact of war‐related grief on mental health, the samples consisted of bereaved and nonbereaved participants. All participants reported exposure to at least one war‐related traumatic event beyond loss of relatives during the war. After a complete description of the study to the subjects, informed consent was obtained. Participants who were 16 or 17 years old (18.9% of the sample) were asked to provide informed consent after consulting one of their parents. The study was approved by the Ethics Committee of the University of Amsterdam. Participants were compensated with 5 Euros for their participation.

Lists of all families who had lost relatives during the war in the selected municipalities were provided by communal authorities. Out of 406 contacted participants, 95 did not meet the inclusion criteria (younger than 6 years old during the war with regard to the subsample of bereaved young adults and having no children with regard to the subsample of single mothers). Of 311 potential participants, 27 declined to participate in the study, resulting in a participation rate of 91.3%: 284 of 311. Missing values resulted in the exclusion of five participants; thus, 279 bereaved participants were included in the analyses.

Nonbereaved civilian war survivors were contacted in the same localities as bereaved participants. This group was recruited using a random walk approach (Priebe et al., [23] ) that involved random identification of streets in each locality where bereaved participants were contacted. All participants reported at least one potential war‐related traumatic event. Out of 494 households contacted, 186 did not meet inclusion criteria (younger than 6 years old during the war with regard to the subsample of nonbereaved young adults, having no children with regard to the subsample of married mothers, and loss of first‐degree relatives after the war). Of the remaining 308 potential participants, 21 declined to participate in the study (i.e., a participation rate of 93.2%). Missing values resulted in the exclusion of six participants, resulting in a total sample size of 281 nonbereaved participants.

The mean age of participants in Study 1 was 30.56 years (SD = 14.61, range 16–69) and 75.4% were female. On average, they reported 9.55 years of school (SD = 3.84) and 47.0% of them were unemployed. Participants reported a mean of 17.59 (SD = 13.52) war‐related traumatic events, a mean of 0.16 (SD = 0.80) prewar events, and a mean of 0.16 (SD = 1.04) postwar events.

Sociodemographic characteristics of the participants were assessed on a brief structured questionnaire. Traumatic events were measured using a checklist that assessed 18 war‐related traumatic events (Morina, Rushiti, Salihu, & Ford, [21] ).

MDE and anxiety disorders were assessed with the MINI International Neuropsychiatric Interview that has shown similar diagnostic sensitivity compared to the Structured Clinical Interview for DSM‐III‐R (Sheehan et al., [25] ). The MINI is the only structured psychiatric interview translated for use among Kosovar Albanians (Morina, [18] ). There is lack of information regarding psychometric properties of the Albanian version of the MINI, apart from the assessment of the interrater reliability among interviewers of a prior project across two sessions that was reported to be 90.2% (Priebe et al., [23] ). To culturally adapt the instrument, the authors had applied an earlier version of the instrument to civilian war survivors in Kosovo and subsequently adjusted the translation based on intense discussions with the local interviewers who had applied the MINI (Morina, [18] ). This interview had been used several times in the Kosovar population (Morina et al., [21] ; Priebe et al., [23] ).

PTSD symptoms experienced in the past month were assessed with the Posttraumatic Stress Diagnostic Scale (PDS; Foa, Cashman, Jaycox, & Perry, [10] ) that assesses the 17 PTSD symptoms specified in the DSM‐IV and thereby also the six PTSD symptoms proposed in the ICD‐11. Items are scored on a 4‐point scale ranging from 0 = never to 3 = 5 times per week or more/nearly always. All participants indicated the war‐related traumatic event that bothered them the most prior to filling out the PDS in relation to that event. Foa et al. ([10] ) reported good reliability and concurrent validity of the PDS with other PTSD measures, and that it has satisfactory agreement (κ =.65, agreement = 82%, sensitivity =.89, specificity =.75) with the PTSD module of the Structured Clinical Interview for DSM‐IV (SCID; First, Spitzer, Gibbon, & Williams, [9] ). In this study, the internal consistency of the Albanian version of the PDS (Morina, Bohme, Morina, & Asmundson, [19] ) was α =.95.

Veterans were predominantly recruited from the United Kingdom Service Personnel and Veterans Agency whose medical staff sent invitation letters for research participation to groups of veterans receiving pensions for PTSD or for physical disabilities only. The explicit focus of the study was on veterans’ general well‐being. Of the veterans at confirmed addresses, 51% agreed to an interview. Inclusion criteria were being in receipt of a war pension either for DSM‐IV PTSD or for a physical disability as a result of military service, being younger than 60 years, and having experienced a traumatic event during military service. Of 191 interviewed veterans, 13 did not meet these inclusion criteria and other veterans were omitted on the basis of our exclusion criteria: presence of ongoing PTSD from before military service (1); other severe psychiatric disorder (3); inability to remember key symptom, trauma, and date information (12); unwillingness to disclose such information (9); and a symptom presentation that could not be reliably classified as either an immediate or a delayed onset (11; see Andrews, Brewin, Stewart, Philpott, & Hejdenberg, [4] , for full details). The current report was based on 103 veterans diagnosed on the basis of our research interview with PTSD at some time during the study period and 39 controls receiving war pensions for physical disorders who did not report PTSD at any time. The research was approved by Research Ethics Committees at Royal Holloway and University College London. After complete description of the study to the subjects, written informed consent was obtained.

The mean age of the veterans was 36.27 years (SD = 4.99, range 22–59) and 97.2% were male. Education up to the age of 16 years only was reported by 58.2%, up to 18 years by 25.5%, and degree level or above by 16.3%. The percentage of commissioned officers was 4.2%, noncommissioned officers 38.0%, and other ranks 57.8%. The average length of time they had served in the armed forces was 8.50 years (SD = 4.47). Participants reported a mean of 3.20 (SD = 1.55) traumatic events.

At the start of the clinical interview, veterans were asked to list traumatic service events meeting Criterion A1 of the SCID‐IV, describe and date their first and last service trauma, and identify and date a trauma that they believed had affected them most (the main trauma). Symptoms were then related to these events. Retrospective DSM‐IV diagnoses of current PTSD and thereby also the proposed ICD‐11 diagnoses of PTSD, anchored to the time the participant was discharged from the armed services, were determined during our research interviews using the SCID for DSM‐IV (First et al., [9] ). As in previous studies of military samples (Bliese, Wright, Adler, Castro, & Hoge, [5] ), PTSD Criterion A2 (reporting fear, helplessness, or horror at the time of the event) was not required.

Interviewers were trained to use the SCID and all interviews were audiotaped and transcribed verbatim. Interrater reliability was calculated on a subset of 27 interviews from the full sample of 191 and found to be acceptable: classification as immediate versus delayed‐onset versus no DSM‐IV PTSD (96% agreement, κ =.94).

First, sociodemographic and trauma‐related characteristics were summarized using means and standard deviations or percentages, as applicable and available for each sample. Second, we calculated the proportions of participants in each sample meeting the PTSD diagnosis following the DSM‐IV and proposed ICD‐11 criteria, respectively. Two‐tailed binomial approximation z tests for proportions were used to evaluate differences in proportions. Third, we calculated the odds ratios and their 95% confidence intervals of comorbid MDE and anxiety disorders other than PTSD among civilian war survivors meeting only the DSM‐IV criteria or the proposed ICD‐11 criteria for Study 1. The numbers were too small to do this analysis for Study 2.

In Study 1, the PTSD prevalence according to ICD‐11 criteria was not significantly different from the PTSD prevalence based on the DSM‐IV (z = 1.66, p =.10). As shown in Table [NaN] , 48 participants (8.6%) met DSM‐IV criteria only (and thus not ICD‐11 criteria) and 22 participants (3.9%) met the proposed ICD‐11 criteria and not the DSM‐IV criteria, leaving 87.5% with an unchanged diagnostic status. An examination of the PTSD symptom clusters revealed mixed results. As compared to DSM‐IV criteria, significantly fewer individuals met ICD‐11 criteria for intrusion and significantly more individuals met ICD‐11 criteria for avoidance (both ps <.01). There was no significant difference between both sets of criteria with regard to hyperarousal (p =.91).

Proportion Meeting DSM‐IV and Proposed ICD‐11 PTSD Criteria, Change From DSM‐IV Status, and Test of Comparisons

• 7 Note
• 8 DSM‐IV = Diagnostic and Statistical Manual of Mental Disorders (4th ed.); ICD‐11 = International Classification of Diseases (11th ed.); PTSD = posttraumatic stress disorder.
• 9 *p <.05. **p <.01. ***p <.001.

In Study 2, the prevalence of PTSD was not significantly different regarding the two criteria sets (see Table [NaN] ). Three participants (2.1%) met DSM‐IV criteria only (and thus not ICD‐11 criteria) and nine participants (6.3%) met the proposed ICD‐11 criteria and not the DSM‐IV criteria, leaving 91.5% with an unchanged diagnostic status. As compared to DSM‐IV criteria, significantly more individuals met ICD‐11 criteria for avoidance (p <.05). There was no significant difference between both sets of criteria with regard to intrusion or hyperarousal (both ps >.10). See Table [NaN] for the proportion of participants reporting single DSM‐IV and proposed ICD‐11 PTSD items.

Proportion Reporting DSM‐IV and Proposed ICD‐11 PTSD Items

• 10 Note
• 11 DSM‐IV = Diagnostic and Statistical Manual of Mental Disorders (4th ed.); ICD‐11 = International Classification of Diseases (11th ed.); PTSD = posttraumatic stress disorder.

Table [NaN] presents rates of comorbidity of PTSD with MDE and anxiety disorders among Study 1 participants meeting PTSD criteria according to both criteria sets. An examination of participants meeting DSM‐IV criteria only and participants meeting ICD‐11 criteria only revealed that rates of comorbidity were generally lower in the ICD‐11 group, but differed significantly only with respect to MDE, with fewer participants in this group reporting the diagnosis (p =.01). For Study 2, numbers of participants meeting PTSD criteria according to both criteria sets were too small to conduct a corresponding analysis.

Proportion of Civilian War Survivors With PTSD Meeting Criteria for Comorbid Depression or Anxiety Disorders

• 12 Note
• 13 PTSD = posttraumatic stress disorder; DSM‐IV = Diagnostic and Statistical Manual of Mental Disorders (4th ed.); ICD‐11 = International Classification of Diseases (11th ed.); MDE = major depressive episode; PD = panic disorder; AG = agoraphobia; SP = social phobia; OCD = obsessive compulsive disorder; GAD = generalized anxiety disorder; AD = anxiety disorder. Odds ratios (ORs) compared individuals only meeting ICD criteria for PTSD to individuals only meeting DSM‐IV criteria for PTSD.

Our results indicate that the new approach proposed for ICD‐11 need not make a substantial difference to PTSD prevalence. This finding held across two samples that differed markedly in gender and composition (civilian vs. military), and are consistent with van Emmerik and Kamphuis ([27] ). This finding is in some ways surprising given the attempt to define the core features of the disorder more narrowly, requiring a minimum of only three symptoms. Although fewer symptoms are required for the diagnosis, the corresponding reduction in the number of qualifying symptoms (from 17 to 6) appears to have pre‐empted any tendency to inflate diagnosis rates. Our findings are in line with evaluations of other proposals to reduce the number of PTSD symptoms (Spitzer et al., [26] ), which have similarly found few changes in prevalence (Grubaugh, Long, Elhai, Frueh, & Magruder, [13] ) and suggest that the 17 symptoms described in the DSM‐IV are not all required to assess PTSD.

One effect of the ICD‐11 proposals, which was observed in both datasets, was to decrease the number of individuals meeting the reexperiencing criterion. This was to be expected given that repetitive intrusive memories, once thought to be unique to PTSD, are now known to be found in many other disorders (Brewin, Gregory, Lipton, & Burgess, [7] ). Moreover, under the DSM‐IV reexperiencing symptoms included intrusive thoughts about the trauma, which may be more characteristic of depressive rumination (Friedman et al., [12] ). Among civilian war survivors, more participants met the DSM‐IV reexperiencing criterion than the ICD‐11 criterion (i.e., distressing dreams and flashbacks). This was mainly due to the finding that these two symptoms were the two least prevalent reexperiencing symptoms. As this did not lead to a significant difference on the prevalence of overall PTSD (87.5% of individuals in the civilian sample who met ICD‐11 criteria for PTSD also fulfilled DSM‐IV criteria for PTSD), however, this result might support the notion that distressing dreams and flashbacks represent the core reexperiencing symptoms. Thus, individuals reporting either distressing dreams or flashbacks were very likely to also meet the other PTSD criteria. Among war veterans, however, there was no difference between the DSM‐IV and ICD‐11 prevalence of the reexperiencing criterion. In this sample, 54.2% of individuals reported distressing dreams, which is double as much as in the civilian sample and might suggest that distressing dreams might be more characteristic of combat trauma. Yet, the difference might also be a result of different measurements (PDS vs. SCID).

Perhaps the most substantial effect of the changes was to increase the number of individuals meeting the avoidance/numbing criterion. Our data support this notion; in both samples the number of individuals meeting the avoidance criterion was higher for the ICD‐11 system than for the DSM‐IV system. This is most plausibly a result of the reduced threshold whereby three out of seven avoidance symptoms must be present to meet the DSM‐IV avoidance criterion as compared to one out of two symptoms following the ICD‐11. One potential implication is that previously a substantial proportion of individuals diagnosed with PTSD showed active avoidance of internal or external stimuli that reminded them of their trauma, but failed to meet the criterion because other symptoms in this cluster were lacking. Substantial literature shows that the active avoidance symptoms are functionally distinct from the other Criterion C symptoms in DSM‐IV (e.g., social withdrawal, loss of interest in activities; Yufik & Simms, [29] ). Recent analyses support the position that requiring active avoidance (proposed both for ICD‐11 and DSM‐5) is likely to refine the PTSD diagnosis and reduce spurious diagnoses being applied to individuals who are primarily suffering from depression (Forbes et al., [11] ).

Relatedly, the reduced set of symptoms led to significantly less comorbidity with major depression. The removal of symptoms, a number of which are now known to be characteristic of depression (e.g., intrusive memories, distressing recollections, social withdrawal, foreshortened future), may have improved the specificity of the remaining symptoms sufficiently to reduce comorbidity, as was the intention. Alternatively, the reduced comorbidity with major depression may be viewed as a correlate of lower symptom severity in participants meeting the proposed ICD‐11 criteria for PTSD compared to participants meeting the DSM‐IV criteria. Interpretation of our findings will differ depending on whether the ICD‐11 is perceived as having a more lenient avoidance/numbing criterion as compared to the DSM‐IV or whether these symptoms are perceived as being part of a separate condition.

As previously noted, however, comorbidity was not reduced by these means in a treatment‐seeking sample (van Emmerik & Kamphuis, [27] ). Yet, this dataset was modest in size, and the groups meeting ICD‐11 and DSM‐IV criteria overlapped substantially. The current analyses use much larger numbers and permit the most appropriate comparison, namely between individuals meeting DSM‐IV but not ICD‐11 criteria, and vice versa. Replication of these findings is nevertheless needed. Other strengths of the study include the use of disparate samples that differ markedly in gender, age, culture, and training for exposure to war trauma and that data resulting from both samples yielded similar results regarding the prevalence of PTSD. This is consistent with growing evidence for the cross‐cultural validity of the PTSD construct (Hinton & Lewis‐Fernandez, [14] ).

Limitations include the focus on war trauma only and the low rates of some comorbid anxiety disorders. Additionally, the PDS used in the civilian sample measures the frequency of PTSD symptoms only and not the level of impairment. Further, the use of a self‐report in the civilian sample represents a less accurate assessment of PTSD symptoms than garnered from the use of clinical interviews. A potential overrepresentation of PTSD symptoms, however, is not likely to have impacted the current results as such an overrepresentation should be present in both formulations of PTSD (i.e., DSM‐IV and ICD‐11). The use of a self‐report in the first sample and a structured interview in the second sample might also represent a strength of this study given the similar results in both samples. In addition, reliance on previously collected data prevented us from testing proposed ICD‐11 against DSM‐5 diagnoses. Although the participation rate in the veteran sample was only around 50%, the specific analyses we have reported do not require a representative or unbiased sample, so this is unlikely to be a serious limitation. Further research is needed to examine to what extent the ICD‐11 criteria and DSM‐IV criteria demarcate different forms of psychopathology, despite yielding comparable prevalences of PTSD. Additionally, as ICD‐11 criteria include only one symptom that lends itself to objective measurement (exaggerated startle response), future research needs to also investigate potential scientific and clinical implications of the reduced number of the required PTSD objective symptoms.

The symptoms selected for the ICD‐11 are those that appear at present to distinguish PTSD most clearly from other disorders: the reexperiencing of the traumatic event(s), the deliberate avoidance of reminders of the traumatic event(s), and certain hyperarousal symptoms. Five of the six symptoms selected to assess PTSD in ICD‐11 were found to be among the most highly predictive of a PTSD diagnosis in the DSM‐IV Field Trial (Kilpatrick et al., [15] ). The ICD‐11 criteria aim at maximizing clinical utility, leading to more reliable diagnosis in nonspecialized settings such as primary care, and more efficient diagnosis after large‐scale traumatic events. If the current results are replicated another benefit may be a reduction in comorbidity with depression, allowing researchers to measure and study a narrower and more clearly defined phenotype corresponding to this particular response to extreme stress. Thus, the ICD‐11 may under certain circumstances offer a useful alternative to the DSM‐5 PTSD diagnosis, which remains true to the DSM‐IV in offering a detailed and comprehensive clinical picture of the disorder for mental health practitioners.