LRSC

LRSC
SHORT COMMUNICATION Social relationship satisfaction and PTSD: which is the chicken and which is the egg? Sara A. Freedman*, Moran Gilad, Yael Ankri, Ilan Roziner $and Arieh Y. Shalev % Center for Traumatic Stress Studies, Hadassah University Hospital, Jerusalem, Israel Background: Impaired social relationships are linked with higher levels of posttraumatic stress disorder (PTSD), but the association’s underlying dynamics are unknown. PTSD may impair social relationships, and, vice versa, poorer relationship quality may interfere with the recovery from PTSD. Objective: This work longitudinally evaluates the simultaneous progression of PTSD symptoms and social relationship satisfaction (SRS) in a large cohort of recent trauma survivors. It also explores the effect of cognitive behavior therapy (CBT) on the association between the two. Method: Consecutive emergency department trauma admissions with qualifying PTSD symptoms (n 501) were assessed 3 weeks and 5 months after trauma admission. The World Health Organization Quality of Life evaluated SRS and the Clinician Administered PTSD Scale evaluated PTSD symptom severity. Ninety-eight survivors received CBT between measurement sessions. We used Structural Equation Modeling to evaluate cross-lagged effects between the SRS and PTSD symptoms. Results: The cross-lagged effect of SRS on PTSD was statistically significant (b 0.12,p 0.01) among survivors who did not receive treatment whilst the effect of PTDS on SRS was nil (b 0.02,p 0.67). Both relationships were non-significant among survivors who received CBT. Discussion: SRS and PTSD are highly associated, and this study shows that changes in SRS in the early aftermath of traumatic events contribute to changes in PTSD, rather than vice versa. SRS impacts natural recovery, but not effective treatment. This study suggests that being satisfied with one’s relationships might be considered as an important factor in natural recovery from trauma, as well as in intervention. Keywords:Social relationship satisfaction;PTSD;natural recovery Responsible Editor: Cherie Armour, University of Ulster, UK. *Correspondence to: Sara A. Freedman, School of Social Work, Bar Ilan University, Ramat Gan, Israel, Email: [email protected] For the abstract or full text in other languages, please see Supplementary files under ‘Article Tools’ Received: 16 June 2015; Revised: 2 November 2015; Accepted: 9 November 2015; Published: 16 December 2015 D ifficulties in social relationships, including rela- tionship quality, satisfaction, intimacy, cohesion, and sexual satisfaction, have all been associated with the presence of posttraumatic stress disorder (PTSD) in one or both partners (Dekel, Enoch, & Solomon, 2008; Koenen, Stellman, Sommer, & Stellman, 2008; Lunney & Schnurr, 2007). This association, although well estab- lished, is not well understood. Social relationships comprise many different factors and can be assessed from different standpoints. One aspect is satisfaction with intimate social relationships, usually marital partners. Another refers to satisfaction with social roles, suchas parenting. A further factor is related to behavior patterns, such as within a marriage. Social support is a particular facet of social relationships, examining perceived readiness of others to provide help in times of need. Clearly, these different factors overlap, such that higher perceived social support from a spouse is likely related to reported satisfac- tion in that relationship, as well as being related to nurturing behaviors within that relationship (Cundiff, Smith, Butner, Critchfield, & Nealey-Moore, 2015). However, most studies have examined these factors separately. Two conflicting hypotheses exist regarding the role of social relationships in PTSD. On the one hand, PTSD $Present address: School of Social Work, Bar Ilan University, Ramat Gan 52900, Israel.%Present address: Department of Psychiatry, New York University School of Medicine, New York, NY, USA. PSYCHO TRAUMATOLOGY EUROPEAN JOURNAL OF European Journal of Psychotraumatology 2015.#2015 Sara A. Freedman et al. This is an Open Access article distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), allowing third parties to copy and redistribute the material in any medium or format, and to remix, transform, and build upon the material, for any purpose, even commercially, under the condition that appropriate credit is given, that a link to the license is provided, and that you indicate if changes were made. You may do so in any reasonable manner, but not in any way that suggests the licensor endorses you or your use. 1 Citation: European Journal of Psychotraumatology 2015,6: 28864 – http://dx.doi.org/10.3402/ejpt.v6.28864 (page number not for citation purpose) may lead to poorer social relationships. There is some support for this notion. First, studies have shown that its particularly numbing symptoms have been shown to be negatively related to relationship satisfaction (Campbell & Renshaw, 2013). Second, two prospective studies with military populations have examined the association of relationships and PTSD, one showing that PTSD results in poorer relationships (Campbell & Renshaw, 2013), and the other that increases in PTSD results in a detrimental effect on marital relationships and parenting (Gewirtz, Polusny, DeGarmo, Khaylis, & Erbes, 2010). Third, studies examining PTSD and social support have shown that initial PTSD levels predict later social support in military populations (e.g., King, Taft, King, Hammond, & Stone, 2006; Solomon & Mikulincer, 1990). These studies seem to indicate therefore that PTSD has detrimental effects on social relationships. The opposing hypothesis that poorer social relationships, which in- crease the likelihood of developing PTSD, also has some support in the literature, because the social support in general has been shown to be a consistent predictor of PTSD (Andrews, Brewin, & Rose, 2003). Studies that have examined PTSD and relationship and role satisfac- tion, and behavior, have not, to our knowledge, examined the possibility that poorer relationships may lead to PTSD. Indeed, the studies described above that examined relationship satisfaction rather than social support, assessed relationship satisfaction only at the second time point, and therefore the hypothesis that PTSD might be sub- sequent to poorer relationships could not be tested. One longitudinal study has shown that both these opposing hypotheses may be correct, such that perceived social support amongst family members is a predictive factor of PTSD soon after a traumatic event, but over time, this relationship is reversed, and PTSD levels then predict social support (Kaniasty & Norris, 1993). A related area of research is the effect that social rela- tionships have on recovery from PTSD. A small number of studies have shown that greater social support (Thrasher, Power, Morant, Marks, & Dalgleish, 2010) and lower ex- pressed emotion (Tarrier, Sommerfield, & Pilgrim, 1999) are related to recovery from chronic PTSD. However, the impact of social relationships has not been considered as a potential factor in early recovery. The first months after trauma exposure are critically important for the devel- opment of persistent PTSD. Indeed, most survivors who meet initial PTSD diagnostic criteria recover from PTSD within 1 year (Freedman, Brandes, Peri, & Shalev, 1999; Kessler, 2000). It is plausible that qualities of interper- sonal relationships may critically affect the likelihood of recovery early after a traumatic event, either sponta- neously or with treatment. To our knowledge, no study has previously examined this relationship. To address this gap, we here present a novel analysis of data from a randomized control study, in which patientswere evaluated at multiple time points during the months that followed trauma exposure (Shalev et al., 2011, 2012). In addition to symptom levels at all time points, subjects’ perceptions of their social relationships were also as- sessed. This measure assesses three of the social relation- ship factors described above: perceived social support, perceived satisfaction with social relationships in general, and perceived satisfaction with intimate relationships. Specifically, we use these data to address two alternative hypotheses: decreased social relationship satisfaction (SRS) would be associated with subsequent higher levels of PTSD, and that increased PTSD would be associated with subsequent poorer SRS, in both cases controlling for treatment type. Method Participants Participants were 501 individuals who had attended the emergency room following a civilian trauma meeting Criterion A of DSM (American Psychiatric Association, 2000). Women were 49.7% of the sample, men were 50.3%. The mean participants’ age was 36.22 (SD 11.84) and it ranged from 20 to 69. SES was measured on a 5-point scale, ranging from very low (1) to very high (5); the mean SES was 2.4 (SD 1.1). Measures Clinical interviews Clinician Administered PTSD Scale (CAPS; Blake et al., 1995): PTSD was assessed at both time points using the CAPS. The CAPS gives a score for both frequency and intensity of the 17 PTSD symptoms (DSM IV), and a continuous score is calculated by summing all these. In this study, Cronbach’sawas 0.93 at Time I and 0.96 at Time II. SCID Structured Clinical Interview for DSM-IV, (First, Spitzer, Gibbon, & Williams, 1997): This was used to screen for all current and past Axis I disorders. In this data set,past depressionwas measured using the SCID at Clinical Interview I. It was dummy-coded with 1 yes. SRS was measured using theWorld Health Organiza- tion Quality of Life-BREF (Group, 1998): This is the 26-item version of the original questionnaire and mea- sures subjective quality of life, in four dimensions: phys- ical health, psychological health, social relationships, and environment. Responses are measured on a 5-point scale, with a higher number indicating a better perceived quality of life. In this study, the three questions from the social relationships section were included: How satisfied are you with your personal relationships?; How satisfied are you with your sex life?; and How satisfied are you with the support you get from your friends? In this study, Sara A. Freedman et al. 2(page number not for citation purpose) Citation: European Journal of Psychotraumatology 2015,6: 28864 – http://dx.doi.org/10.3402/ejpt.v6.28864 Cronbach’sawas 0.75 at Clinical Interview I and 0.81 at Clinical Interview II. Negative life events(NLE) were measured using the Stressful Life Events Screening Questionnaire*SLESQ (Goodman, Corcoran, Turner, Yuan, & Green, 1998); this is the Hebrew version of the SLESQ, used in several previous studies. A total number of endorsed events were used. Sociodemographic variableswere (1) sex (dummy- coded, with male 1); (2) age in years; (3) schooling years; (4) marital status (dummy-coded, with married 1 and otherwise 0); (5) number of children (0 9); (6) household density (defined as the number of household members divided by the number of rooms in the house- hold); and (7) self-reported income (based on partici- pants’ responses to a 5-point scale ranging from 1 way below average to 5 way above average). Trauma type, with two types (MVA and Terror) as the dummy-coded variables, and other types serving as the base. Randomized control trial(RCT) condition was ex- pressed as two dummy-codes of WL and PE/CT, with non-RCT serving as the base. Procedure This study examined data collected as part of the Jerusalem Trauma Outreach and Prevention Study (J-TOPS; ClinicaltrialsGov Identifier: NCT0014690); the study included systematic outreach and follow-up of recent trauma survivors, as well as an embedded rando- mized controlled study examining the effects of early interventions for preventing chronic PTSD. The study’s procedures have been fully described previously (Shalev et al., 2011) and will be briefly described here. Subjects were adults aged 18 70 who were consecu- tive trauma survivors attending the emergency room of a level I trauma center. Traumatic events were mixed civilian events, with the majority (80%) being motor vehicle accidents. Eligible participants were recruited via a telephone interview 10 days following the event (N 4,743) and gave verbal informed consent. During this initial telephone interview, participants who had experienced a traumatic event according to criteria A1 and A2, and met other study criteria (N 1,996) were assessed for initial symptoms of acute stress disorder. Participants with acute PTSD symptoms (N 1,502) were invited to attend a clinical interview 3 weeks post trauma, of which 756 attended. At this clinical interview (Clinical Interview I), subjects signed written informed consent and were assessed for current and lifetime psy- chiatric disorders using the SCID, and for the presence of PTSD symptoms. Three hundred ninety-seven subjects showed sufficient symptoms (acute PTSD minus the time criterion) and were invited to participate in the rando- mized controlled trial. Two hundred ninety-six subjectsagreed to enter the trial and were randomized using equipoise stratified randomization to one of five treat- ment arms: prolonged exposure (PE,N 63), cognitive therapy (CT,N 40), SSRI (N 23), placebo (N 23), and waiting list control (N 93). Results indicated that PE and CT showed similar effectiveness, and for the pur- poses of these analyses are considered one group, namely PE/CT. All patients received 12 weeks of treatment. As described above, the RCT was embedded in a follow-up study of all subjects, and as such partici- pants (N 756) who attended the first clinical interview were reassessed 5 months after the trauma (N 604, Clinical Interview II), regardless of participation in the RCT. Assessors were blind to participation in the RCT. Hadassah University Hospital’s Institutional Review Board approved and monitored the study. The data used in this analysis concentrate on 501 individuals who have available data at the pre- and post- early treatment clinical assessments; that is, at 3 weeks and 5 months post trauma, Clinical Interviews I and II. Three groups were included: those who received either PE or CT (PE/CT,N 98); those who were in the waiting list control group (WL,N 90); and those who did not enter the RCT and therefore did not receive any treatment (non-RCT,N 313). It is important to include this non- treatment group, because it represents an understudied population, and it is rare to have data that allow for comparison of subjects who did and did not enter an RCT. The SSRI/Placebo group was small and had in- sufficient numbers at follow-up to be included in the present analyses. SRS and PTSD symptom levels were assessed at both time points. Data analytic plan The main analysis approach was Structural Equation Modeling, done with the Mplus 7.11 program (Muthe´n & Muthe´ n, 2012). The minimal covariance coverage in the variance covariance matrix used in the analyses was 0.83. To take advantage of all the available data, models were fit using full-information maximum likelihood estimation with robust standard errors (Little & Rubin, 2003). Following recommendations of Hu and Bentler (1999), we report fit indexes of two types: the Tucker Lewis index (TLI) and the Comparative Fit Index (CFI), and two indexes of misfit: root mean-square error of approx- imation (RMSEA) and standardized root mean-square residual (SRMR). NNFI and CFI close to or above 0.95, combined with RMSEA below 0.06 and SRMR below 0.08, are considered indicative of acceptable fit. SRS was specified as a latent variable, indicated by its three items. PTSD was also specified as a latent variable, measured with three indicators, each created as a random third of the scale items using the accepted approach of parceling (Bandalos, 2002; Stacy, Bentler, & Flay, 1994). To test for reciprocal causal associations between SRS and PTSD, Social relationship satisfaction and PTSD Citation: European Journal of Psychotraumatology 2015,6: 28864 – http://dx.doi.org/10.3402/ejpt.v6.28864 3 (page number not for citation purpose) we fit, within the SEM framework, a cross-lagged panel model (Finkel, 1995). Such models allow for testing the causal associations of two variables while considering for their stabilities and controlling for measuring issues. For this, we allowed for measurement errors of the same indicators to correlate over time in autoregressive paths and fixed factor loadings to equality over time. Results Descriptive statistics of observed research variables and their correlations with relationship satisfaction and PTSD measures are provided in Table 1. As can be seen in Table 1, PTSD symptoms are nega- tively correlated with SRS, at both time points. Higher frequency of NLE and higher levels of past depression are associated with lower SRS at both time points. In addition, age is negatively correlated with SRS and posi- tively correlated with PTSD, both at T2. Participants who reported higher income also reported better RS and lower PTSD at T1, but not at T2. As the first stage of the main analyses, we tested the measurement model. It yielded acceptable results:x 2 (46,N 501) 66.26,p 0.03, TLI 0.995, CFI 0.996, SRMR 0.030, RMSEA 0.030 (90% CI 0.010; 0.045). We proceeded then to test the cross-lagged panel model. To this model, we added the two dummy variables ex- pressing membership in the two RCT experimental groups as predictors of both T1 and T2 relationship satisfaction and PTSD. Parceling out the group membership fromT1 measures allowed us to account for the initial im- balance in the groups’ composition. The group member- ship effects upon the T2 measures controlled for T1 measures allow us to estimate the RCT impact upon the change over time in these measures. We also added to the model, as predictors of each of the four content variables, those sociodemographic and background variables that were correlated with any of the content research variables (Table 1). This structural model fit the data well, with x 2(114,N 501) 208.80,pB0.0001, TLI 0.978, CFI 0.983, SRMR 0.039, and RMSEA 0.041 (90% CI 0.032; 0.049). In this model, all the paths emitted from two control variables, number of children and self- reported income, were not statistically significant. These variables were deleted from the model. The paths from NLE to relationship satisfaction and PTSD at T2 were also non-significant and were therefore fixed to zero. The resulting model (Fig. 1) showed good fit to the data, withx 2(92,N 501) 158.92,pB0.0001, TLI 0.984, CFI 0.988, SRMR 0.037, and RMSEA 0.037 (90% CI 0.027; 0.047). As seen in Fig. 1, both SRS and PTSD exhibited some stability across time (with stability coefficients of 0.62 and 0.56, accordingly). The correlations between them were rather high, both at T1 (r 0.42) and T2 (r 0.50). Importantly, the cross-lagged effect of PTSD upon SRS was close to zero (b 0.02,p 0.67), whereas the parallel effect of SRS upon PTSD was small in magni- tude, but statistically significant (b 0.12,p 0.01). Ta b l e 1 .Means and standard deviations for the research variables and their correlations with SRS and PTSD Variable MeanSDSocial relationship satisfaction T1Social relationship satisfaction T2PTSD T1PTSD T2 RCT group: waiting list a 0.18 0.38 0.09 0.18*** 0.22*** 0.29*** RCT group: PE/CT a 0.20 0.40 0.16*** 0.02 0.24*** 0.01 Sex: male a 0.50 0.50 0.02 0.06 0.02 0.05 Age (in years) 36.22 11.84 0.06 0.16*** 0.04 0.11* Education (in years) 13.19 2.67 0.00 0.00 0.00 0.04 Marital status: married a 0.51 0.50 0.00 0.05 0.02 0.06 Number of children 1.60 1.98 0.04 0.08 0.06 0.11* Household density 1.05 0.59 0.02 0.06 0.07 0.06 Income 2.40 1.10 0.09* 0.08 0.13** 0.01 Negative life events 1.80 1.77 0.13** 0.17*** 0.05 0.07 Past depression a 0.27 0.44 0.10* 0.16*** 0.13** 0.05 Trauma type: terror a 0.13 0.33 0.01 0.02 0.09* 0.05 Trauma type: MVA a 0.83 0.37 0.01 0.02 0.06 0.02 Social relations T1 3.29 0.96* Social relations T2 3.50 0.97 0.58**** PTSD T1 1.58 0.73 0.40*** 0.29**** PTSD T2 0.81 0.74 0.32*** 0.52*** 0.57**** MVA, motor vehicle accident; Terror, terrorist attack. *pB0.05; **pB0.01; ***pB0.001. aDummy-coded variable, 1 the value specified in the variable name, 0 other. Sara A. Freedman et al. 4(page number not for citation purpose) Citation: European Journal of Psychotraumatology 2015,6: 28864 – http://dx.doi.org/10.3402/ejpt.v6.28864 This pattern of results is consistent with the hypothesis that initial levels of SRS are the driving force in the development of PTSD. At the next step, we added to the model the interactive effects of RCT group membership, WL, or PE/CT, and the initial level of each of the focal research variables, SRS or PTSD, upon the other variable at T2. To obtain a clear picture, we tested each of these effects within separate models. Three out of four interactive effects failed to reach statistical significance (for WL PTSD,p 0.66; for PE/CT PTSD,p 0.63; and for WL SRS,p 0.11). The interaction of WL SRS was significant,b 0.19, p 0.02. To illustrate the form of this interaction, the cross-lagged model was fit separately for the three experi- mental groups. The path from T1 relationship satisfaction to T2 PTSD was negative and significant in two groups, non-RCT (b 0.18,pB0.05) and WL (b 0.26, pB0.05), but was not significant in the PE/CT group (b 0.12,p 0.30). Discussion To the best of our knowledge, this is the first prospec- tive analysis examining SRS, PTSD, and treatment re- sponse in a civilian population. The results indicate thatincreased PTSD symptoms are correlated with decreased SRS, and this corroborates previous studies (Campbell & Renshaw, 2013; Gewirtz et al., 2010). However, the results of the path analysis indicate that poorer SRS may drive PTSD symptoms, rather than the reverse. Previous studies (Campbell & Renshaw, 2013; Gewirtz et al., 2010) have shown that PTSD results in decreased satisfaction, but as noted above, these studies did not examine initial relationship satisfaction. These results imply that patients who present with trauma exposure and relationship difficulties may be particularly vulnerable to PTSD development. It is possible that the discrepancy with previous studies arises from the differ- ent populations studied*civilian as opposed to military. As has been noted (De Burgh, White, Fear, & Iversen, 2011), the marital relationship is placed under particular strain in military families, due in part to long absences and frequent moving, and therefore it may not be rele- vant to compare these populations. In survivors who received treatment, the significant relationship between decreased satisfaction with relation- ships and elevated PTSD is not apparent, whereas it remains significant in those groups who did not receive treatment, either within the RCT (waiting list) or those Fig. 1.Structural equation model of cross-lagged relationship satisfaction and PTSD effects with standardized parameters. x2(92,N 501) 158.92,pB0.0001, TLI 0.984, CFI 0.988, SRMR 0.037, and RMSEA 0.037. The solid lines indicate paths statistically signi cant atpB0.05. The dotted lines indicate non-signi cant paths. NLE, negative life effects; SRS, social relationship satisfaction; WL, waiting list; PE/CT, prolonged exposure/cognitive therapy. Social relationship satisfaction and PTSD Citation: European Journal of Psychotraumatology 2015,6: 28864 – http://dx.doi.org/10.3402/ejpt.v6.28864 5 (page number not for citation purpose) who did not enter the treatment trial. These results do not support previous studies showing that greater social support is related to better treatment response (Thrasher et al., 2010). This may be explained by the timing of assessments in the current study. Previous studies have all examined the treatment or effects of chronic PTSD, rather than preventative treatment early after a traumatic event. Patients in the current study had suffered from symptoms for a relatively short period of time (5 weeks) before beginning treatment. This may not be long enough for the adverse effects of PTSD on previously satisfying relationships to be felt, and therefore effects on treatment response may not be seen. The current results may indicate that the process of natural recovery is enhanced when the individual reports satisfaction with relationships, but impaired with poorer satisfaction. Treatment appears to amend this relation- ship, perhaps showing that when an intervention is successful at impacting PTSD levels, the levels of relationship satisfaction is immaterial to outcome. This may also be attributed to a therapeutic relationship that compensates for the lack of satisfying social relationships (Beutler, Forrester, Holt, & Stein, 2013). Taken together, these results indicate that relationship satisfaction may play a part in the development of PTSD, and natural recovery from it. Chronic PTSD is associated with marital difficulties and parenting problems (Dekel & Monson, 2010) which in themselves can lead to trauma- tization of other family members (Berz, Taft, & Watkins, 2008). The results presented here may show that even though early treatment has little long-term benefit over later treatment (Shalev et al., 2012) it may positively impact family relationships and this has beneficial effect beyond PTSD symptoms. This study is limited first by the measurement of relationship satisfaction. Recent studies examining family interactions have preferred to use multiple assessments, thus gaining input, for example, from both partners in a relationship (Holmbeck, Li, Schurman, Friedman, & Coakley, 2002) as opposed to self-report measures of one partner. The second limitation regards the sample, consisting of one-time traumatic events in a civilian population. This analysis should be replicated in other trauma populations. Conclusions These novel results indicate that the impact of PTSD and social relationships on each other needs to be more fully explored. Recent clinical trials have shown the impor- tance of including significant others in PTSD treatment (Monson et al., 2012), and these results support this approach. Future studies might systematically assess social relationships, their impact on treatment, and the impact that treatment (or lack of it) has on them. Conflict of interest and funding Dr. Shalev received an investigator-initiated grant from Lundbeck Pharmaceuticals Ltd. (Denmark) for this study and for an ongoing collaborative study (PI Dr. Joseph Zohar) entitled ‘‘Prevention of PTSD by Escita- lopram.’’ Other co-authors have no conflicts of interest to declare. This study was supported by a research grant # MH071651 from the NIMH. References American Psychiatric Association (2000).Diagnostic and statistical manual of mental disorders(4th ed., text rev.). Washington, DC: Author. Andrews, B., Brewin, C. R., & Rose, S. (2003). Gender, social support, and PTSD in victims of violent crime.Journal of Traumatic Stress,16, 421 427. doi: http://dx.doi.org/10.1023/ A:1024478305142 Bandalos, D. L. (2002). The effects of item parceling on goodness- of- t and parameter estimate bias in structural equation modeling.Structural Equation Modeling,9,78 102. doi: http:// dx.doi.org/10.1207/S15328007SEM0901_5 Berz, J. B., Taft, C. T., & Watkins, L. E. (2008). Psychological associations between PTSD symptoms and parenting satis- faction in a female veteran sample.Journal of Psychological Trauma,7(1), 37 45. Beutler, L. E., Forrester, B., Holt, H., & Stein, M. (2013). Common, speci c, and cross-cutting psychotherapy interventions. Psychotherapy (Chicago, Ill.),50, 298 301. doi: http://dx.doi. org/10.1037/a0032405 Blake, D. D., Weathers, F. W., Nagy, L. M., Kaloupek, D. G., Gusman, F. D., Charney, D. S., et al. (1995). The development of a clinician-administered PTSD scale.Journal of Traumatic Stress,8,75 90. doi: http://dx.doi.org/10.1007/BF02105408 Campbell, S. B., & Renshaw, K. D. (2013). PTSD symptoms, disclosure, and relationship distress: Explorations of mediation and associations over time.Journal of Anxiety Disorders,27(5), 494 502. doi: http://dx.doi.org/10.1016/j.janxdis.2013.06.007 Cundiff, J. M., Smith, T. W., Butner, J., Critch eld, K. L., & Nealey- Moore, J. (2015). Af liation and control in marital interaction: Interpersonal complementarity is present but is not associated with affect or relationship quality.Personality & Social Psychology Bulletin,41(1), 35 51. doi: http://dx.doi.org/10. 1177/0146167214557002 De Burgh, H. T., White, C. J., Fear, N. T., & Iversen, A. C. (2011). The impact of deployment to Iraq or Afghanistan on partners and wives of military personnel.International Review of Psychiatry (Abingdon, England),23, 192 200. doi: http://dx. doi.org/10.3109/09540261.2011.560144 Dekel, R., Enoch, G., & Solomon, Z. (2008). The contribution of captivity and post-traumatic stress disorder to marital adjustment of Israeli couples.Journal of Social and Personal Relationships,25(3), 497 510. doi: http://dx.doi.org/10.1177/ 0265407508090870 Dekel, R., & Monson, C. M. (2010). Military-related post-traumatic stress disorder and family relations: Current knowledge and future directions.Aggression and Violent Behavior,15(4), 303 309. doi: http://dx.doi.org/10.1016/j.avb.2010.03.001 Finkel, S. E. (1995).Causal analysis with panel data. Quantitative applications in the social sciences(Vol. 105). Retrieved February 10, 2015, from http://books.google.co.jp/books?hl ja&lr &id WJI5iN-AwHEC&oi fnd&pg PR4&dq Causal analysis with panel data&ots Nut_V29qeb&sig zmR-m-z4DY8eq Sara A. Freedman et al. 6(page number not for citation purpose) Citation: European Journal of Psychotraumatology 2015,6: 28864 – http://dx.doi.org/10.3402/ejpt.v6.28864 cv0j13-WMpHp2o&redir_esc y#v onepage&q Causal analysis with panel data&f false First, M. B., Spitzer, R. L., Gibbon, M., & Williams, J. B. W. (1997). Structured Clinical Interview for DSM-IV Axis I Disorders, Clinician Version (SCID-CV) for DSMIV. Retrieved November 13, 2014, from http://scholar.google.es/scholar?q related:H5J6D 58b8soJ:scholar.google.com/&hl ca&as_sdt 0,5#0 Freedman, S. A., Brandes, D., Peri, T., & Shalev, A. (1999). Predictors of chronic post-traumatic stress disorder. A pro- spective study.The British Journal of Psychiatry: The Journal of Mental Science,174, 353 359. Gewirtz, A. H., Polusny, M. A., DeGarmo, D. S., Khaylis, A., & Erbes, C. R. (2010). Posttraumatic stress symptoms among National Guard soldiers deployed to Iraq: Associations with parenting behaviors and couple adjustment.Journal of Con- sulting and Clinical Psychology,78(5), 599 610. doi: http://dx. doi.org/10.1037/a0020571 Goodman, L. A., Corcoran, C., Turner, K., Yuan, N., & Green, B. L. (1998). Assessing traumatic event exposure: General issues and preliminary ndings for the stressful life events screening questionnaire.Journal of Traumatic Stress,11, 521 542. doi: http://dx.doi.org/10.1023/A:1024456713321 Group, T. W. (1998). Development of the World Health Org- anization WHOQOL-BREF quality of life assessment. The WHOQOL Group. Psychological Medicine,28, 551 558. doi: http://dx.doi.org/10.5.12 Holmbeck, G. N., Li, S. T., Schurman, J. V., Friedman, D., & Coakley, R. M. (2002). Collecting and managing multisource and multimethod data in studies of pediatric populations. Journal of Pediatric Psychology,27,5 18. doi: http://dx.doi. org/10.1093/jpepsy/27.1.5 Hu, L. T., & Bentler, P. M. (1999). Cutoff criteria for t indices in covariance structure analysis: Conventional criteria versus new alternatives.Structural Equation Modeling,6,1 55. doi: http:// dx.doi.org/10.1080/10705519909540118 Kaniasty, K., & Norris, F. H. (1993). A test of the social support deterioration model in the context of natural disaster.Journal of Personality and Social Psychology,64(3), 395 408. doi: http://dx.doi.org/10.1037/0022-3514.64.3.395 Kessler, R. C. (2000). Posttraumatic stress disorder: The burden to the individual and to society.Journal of Clinical Psychiatry, 61(Suppl 5), 4 12; discussion 13 4. King, D., Taft, C., King, L., Hammond, C., & Stone, E. (2006). Directionality of the association between social support and posttraumatic stress disorder: A longitudinal investigation. Journal of Applied Social Psychology,36(12), 2980 2992. doi: http://dx.doi.org/10.1111/j.0021-9029.2006.00138.xKoenen, K. C., Stellman, S. D., Sommer, J. F., & Stellman, J. M. (2008). Persisting posttraumatic stress disorder symptoms and their relationship to functioning in Vietnam veterans: A 14- year follow-up.Journal of Traumatic Stress,21,49 57. doi: http://dx.doi.org/10.1002/jts.20304 Little, R. J. A., & Rubin, D. B. (2003). Statistical analysis with missing data.Sociological Methods Research,18(2 3), 292 326. Lunney, C. A., & Schnurr, P. P. (2007). Domains of quality of life and symptoms in male veterans treated for posttraumatic stress disorder.Journal of Traumatic Stress,20, 955 964. doi: http:// dx.doi.org/10.1002/jts.20269 Monson, C. M., Fredman, S. J., Macdonald, A., Pukay-martin, N. D., Resick, P. A., & Schnurr, P. P. (2012). Effect of cognitive- behavioral couple therapy for PTSD: A randomized controlled trial.JAMA: The Journal of the American Medical Association, 308, 700 709. doi: http://dx.doi.org/10.1001/jama.2012.9307 Muthe´ n, L., & Muthe´ n, B. (2012).Mplus user’s guide(5th ed.). Los Angeles, CA: Author. Shalev, A. Y., Ankri, Y., Israeli-Shalev, Y., Peleg, T., Adessky, R., & Freedman, S. (2012). Prevention of posttraumatic stress disorder by early treatment: Results from the Jerusalem Trauma Outreach and Prevention study.Archives of General Psychiatry,69(2), 166 176. doi: http://dx.doi.org/10.1001/arch- genpsychiatry.2011.127 Shalev, A. Y., Ankri, Y. L. E., Peleg, T., Israeli-Shalev, Y., & Freedman, S. (2011). Barriers to receiving early care for PTSD: Results from the Jerusalem trauma outreach and prevention study.Psychiatric Services,62(7), 765 773. doi: http://dx.doi. org/10.1176/appi.ps.62.7.765 Solomon, Z., & Mikulincer, M. (1990). Life events and combat related PTSD: The intervening role of locus of control and social support.Military Psychology,2(4), 241 256. Stacy, A. W., Bentler, P. M., & Flay, B. R. (1994). Attitudes and health behavior in diverse populations: Drunk driving. Alcohol use, binge eating, marijuana use, and cigarette use.Health Psychology: Of cial Journal of the Division of Health Psychol- ogy, American Psychological Association,13,73 85. doi: http:// dx.doi.org/10.1037/0278-6133.13.1.73 Tarrier, N., Sommer eld, C., & Pilgrim, H. (1999). Relatives’ expressed emotion (EE) and PTSD treatment outcome. Psychological Medicine,29(4), 801 811. Thrasher, S., Power, M., Morant, N., Marks, I., & Dalgleish, T. (2010). Social support moderates outcome in a randomized controlled trial of exposure therapy and (or) cognitive re- structuring for chronic posttraumatic stress disorder.Canadian Journal of Psychiatry,55, 187 190. Social relationship satisfaction and PTSD Citation: European Journal of Psychotraumatology 2015,6: 28864 – http://dx.doi.org/10.3402/ejpt.v6.28864 7 (page number not for citation purpose) R epro duce d w ith p erm is sio n o f th e c o pyrig ht o w ner. F urth er r e pro ductio n p ro hib ite d w ith out p erm is sio n.
LRSC
Relations Among Social Support, PTSD Symptoms, and Substance Use in Veterans Daniel F. Gros, Julianne C. Flanagan, Kristina J. Korte, Adam C. Mills, Kathleen T. Brady, and Sudie E. Back Ralph H. Johnson Veterans Affairs Medical Center, Charleston, South Carolina, and Medical University of South Carolina Social support plays a significant role in the development, maintenance, and treatment of posttraumatic stress disorder (PTSD). However, there has been little investigation of social support with PTSD and its frequent comorbid conditions and related symptoms. Substance use disorders (SUDs) are 1 set of conditions that have yet to be investigated in combination with PTSD and social support. As compared with civilians, veterans are at increased risk for developing both PTSD and SUD. In this study, veterans (N 171) with symptoms of PTSD (76% met diagnostic criteria) and SUD (83% met diagnostic criteria for any dependence) were recruited and completed clinician-rated and self-report measures of PTSD, SUD, and social support. Overall, low social support was reported in the sample. When controlled for the other disorder’s symptoms, PTSD symptoms demonstrated a significant negative relation and SUD symptoms demonstrated a significant positive relation to social support. The PTSD findings are consistent with previous studies on PTSD and social support without SUD comorbidity. However, the SUD findings are inconsistent with previous studies, which focused primarily on older veterans. Together, these findings highlight the significance of social support in individuals with PTSD and SUD and promote future research within comorbid presentations. Keywords:posttraumatic stress disorder, social support, alcohol, veterans Posttraumatic stress disorder (PTSD) is a chronic, debilitating disorder associated with significant distress and impairment. PTSD is the most common mental health disorder among veterans, with approximately 15% of veterans meeting current diagnostic criteria (Seal, Bertenthal, Miner, Sen, & Marmar, 2007). The presence of comorbid substance use disorders (SUDs) with PTSD also is a substantial health concern among veterans. SUD co-occurs with PTSD among approximately 40% of veterans with PTSD (Petra- kis, Rosenheck, & Desai, 2011;Pietrzak, Goldstein, Southwick, & Grant, 2011), and those with co-occurring PTSD and SUD suffer a more complicated course of treatment and less favorable treat-ment outcomes compared with individuals with either disorder alone (Back, 2010;Back, Waldrop, & Brady, 2009;Cohen & Hien, 2006;McCauley, Killeen, Gros, Brady, & Back, 2012). Given the high prevalence and distress associated with comorbid PTSD and SUD, efforts aimed at identifying potential protective factors are especially important for advancing the prevention and treatment of the complex combination of these two conditions. Social support is a potentially important feature in understand- ing how to prevent or treat PTSD. Findings consistently demon- strate that limited social support is associated with more severe PTSD symptoms (Brewin, Andrews, & Valentine, 2000) as well as more severe impairment and suicidal ideation (DeBeer, Kimbrel, Meyer, Gulliver, & Morissette, 2014). Social support also is pos- ited to be a key mechanism in the prevention and treatment of PTSD (Whealin, DeCarvalho, & Vega, 2008). Literature also has demonstrated that social support is a diverse construct in empirical measurement and its manifestations in day-to-day life. For exam- ple, literature indicates that social support available from different individuals in one’s social network (e.g., intimate partners, family members, military unit members and friends) may be differentially influential on symptoms and treatment engagement (Laffaye, Cavella, Drescher, & Rosen, 2008;Pietrzak, Johnson, Goldstein, Malley, & Southwick, 2009). In addition to the source of social support, specific forms of social support, such as positive social interactions, are negatively associated with pretreatment PTSD symptom severity whereas high perceived emotional support is positively associated with increased PTSD treatment response (Price, Gros, Strachan, Ruggiero, & Acierno, 2013). Social support, including the lack thereof, is a salient correlate of SUD treatment engagement and outcome (Manuel, McCrady, Epstein, Cook, & Tonigan, 2007;McCrady, 2004;Zywiak, Long- This article was published Online First October 27, 2016. Daniel F. Gros, Julianne C. Flanagan, Kristina J. Korte, Adam C. Mills, Kathleen T. Brady, and Sudie E. Back, Mental Health Service, Ralph H. Johnson Veterans Affairs Medical Center, Charleston, South Carolina, and Department of Psychiatry and Behavioral Sciences, Medical University of South Carolina. This research was supported by the National Institute on Drug Abuse (NIDA) grant DA030143 (Principal Investigator [PI]: S. E. B.), the De- partment of Veteran Affairs Clinical Science Research and Development Career Development Award CX000845 (PI: D. F. G.), the National Insti- tute on Child Health and Human Development and the Office of Research on Women’s Health grant K12HD055885 (PI: J. C. F.), and by the National Institute of Alcohol Abuse and Alcoholism grant T32AA007474 (K. J. K.). The views expressed in this article are those of the authors and do not necessarily reflect the position or policy of NIDA, the Department of Veterans Affairs, or the U.S. government. Correspondence concerning this article should be addressed to Daniel F. Gros, Mental Health Service 116, Ralph H. Johnson VAMC, 109 Bee Street, Charleston, SC 29401. E-mail:[email protected] Psychology of Addictive BehaviorsIn the public domain 2016, Vol. 30, No. 7, 764 –770http://dx.doi.org/10.1037/adb0000205 764 abaugh, & Wirtz, 2002). Individuals with PTSD and co-occurring SUD are more likely than individuals without SUD to have nu- merous health and psychosocial complications, and lower social support has been documented in those with PTSD and co- occurring disorders than in individuals with a single diagnosis (Blanco et al., 2013;Bowe & Rosenheck, 2015;Campbell et al., 2007;Kaier, Possemato, Lantinga, Maisto, & Ouimette, 2014; Pietrzak et al., 2011). Indeed, the nature of PTSD and SUD symptomatology; the complicating factors that accompany comor- bidity; and associated behaviors such as poor communication related to emotional numbing, aggressive behavior secondary to hyperarousal symptoms, or anger and distrust secondary to chronic substance use behaviors may hinder the availability, and hence the protective utility, of adaptive social support among veterans. How- ever, the research examining the association between PTSD co- morbidities and social support remains limited. One of the few existing studies that examined the effects of PTSD and co-occurring disorders on social support included a large sample (N 1,825) of veterans from Operation Enduring Freedom (OEF)/Operation Iraqi Freedom (OIF). In this study, Brancu and colleagues (2014)found that PTSD was associated with greater distress and lower social support. Veterans with PTSD and a co-occurring mental health disorder did not demonstrate lower social support than veterans with PTSD alone. One factor that may partially explain these findings is the heterogeneity of comorbidities observed in the study, which included a wide variety of mental health diagnoses. Perhaps examining the effects of comorbidity on social support at the disorder level may reveal more distinct patterns of association (e.g., PTSD and SUD). For example,DeBeer and colleagues (2014)examined the role of social support on suicidal ideation using a more homogenous group of individuals with PTSD and comorbid mood disorders. The findings demonstrated that low social support interacted with PTSD and mood symptoms, resulting in greater suicidal ideation than those with higher levels of social support. Given the equivocal findings among the limited existing re- search, there is need for further investigation in this area with a particular focus on common patterns of PTSD comorbidity, such as that with SUD. Developing an improved understanding of these complex associations may facilitate the development and modifi- cation of treatment approaches to enhance social support and thereby improve treatment engagement and outcome among indi- viduals with co-occurring PTSD and SUD. Thus, the purpose of the current study was to address this gap in the literature by examining social support among veterans with PTSD and co- occurring SUD. Given the complicated clinical presentation of this comorbid group, it was predicted that (a) greater severity of PTSD symptoms and (b) greater severity of SUD symptoms would be associated with lower levels of perceived social support. Method Participants Veterans (N 171) seeking treatment for comorbid PTSD and SUD were recruited from Veterans Affairs (VA) treatment clinics, newspaper and Internet advertisements, and flyers posted at local mental health clinics and colleges. Inclusion criteria involved (a) being a veteran, reservist, or member of the National Guard; (b)being 18 – 65 years old; (c) significant symptoms of PTSD and SUD; (d) substance use in the past 90 days; and (e) fluency in English. Exclusion criteria included (a) current or history of psy- chotic symptoms or bipolar affective disorders, (b) current suicidal or homicidal ideation and intent, (c) current eating disorder or dissociative identity disorder, (d) individuals already participating in ongoing PTSD or SUD treatment, and (e) severe cognitive impairment as indicated by a Mini Mental Status Exam score 21. Data were collected as part of an ongoing randomized controlled trial sponsored by the National Institute on Drug Abuse investi- gating the efficacy of an integrated psychosocial treatment for co-occurring PTSD and SUD among veterans (Back et al., 2012). Procedure Potential participants were given a full description of the study procedures and asked to read and sign a consent form approved by the institutional review board before any study procedures or assessments were conducted. The baseline assessment involved semistructured clinical interviews, including the Clinician Admin- istered PTSD Scale (CAPS;Blake et al., 1995) and the Mini International Neuropsychiatric Interview (MINI;Sheehan et al., 1998). Participants also completed the PTSD Checklist-Military (PCL-M;Weathers, Litz, Herman, Huska, & Keane, 1993), the time line follow-back (TLFB;Sobell & Sobell, 1992), and the Deployment Risk and Resiliency Inventory (DRRI;King, King, Vogt, Knight, & Samper, 2006). Measures CAPS.The CAPS is a clinician-rated scale designed to diag- nose current and lifetime PTSD (Blake et al., 1995). The CAPS targets the 17 specific PTSD symptoms from theDiagnostic and Statistical Manual of Mental Disorders, fourth edition (DSM–IV) to assess the intensity and frequency of each symptom on a 5-point Likert scale. The CAPS conducted at baseline was focused on past-month symptoms. Providers of the CAPS attended a 2- to 4-hr CAPS training, watched and corated at least two administrations of the CAPS, administered at least two CAPS under the direct super- vision of a corating supervisor, and demonstrated acceptable in- terrater reliability on their administrations. The CAPS has been shown to have adequate internal consistency, interrater reliability on the same interview, and test–retest reliability over different interviewers (Orsillo, 2002). The internal consistency in the pres- ent study was .91. DRRI.The DRRI consists of 13 subscales to assess prede- ployment, active duty, and postdeployment factors in recently returning combat veterans (King et al., 2006). For the current study, the social support subscale was of interest—the DRRI-L (Post-Deployment Support; items include “I am carefully listened to and understood by family members or friends” and “Among my friends or relatives, there is someone I go to when I need good advice”). Work with veterans has shown the DRRI to demonstrate acceptable internal consistency for the subscales ( s .81) and convergent and discriminative validity (Vogt, Proctor, King, King, & Vasterling, 2008). The internal consistency in the present study was .74. MINI.The MINI is a clinician-rated structured diagnostic interview designed to provide a brief, but accurate, assessment of 765 PAIN, PTSD, AND SUBSTANCE USE a wide range ofDSM–IVpsychiatric disorders, including mood and anxiety disorders as well as SUDs (Sheehan et al., 1998). The MINI was used to assess all of its targeted disorders with the exception of PTSD, which was assessed via the CAPS. Similar training procedures were used for the MINI as were used for the CAPS. The MINI has demonstrated adequate interrater and test– retest reliability across most disorders and has specifically shown good interrater reliability with other structured diagnostic inter- views (Sheehan et al., 1998). PCL-M.The PCL-M is a 17-item self-report measure de- signed to assess PTSD symptom severity related to military/ combat-related trauma (Weathers et al., 1993). Respondents are presented with 17 specific symptoms of PTSD and asked to rate “how much you have been bothered by that problem in the last month” on a 5-point Likert scale. A score 50 or greater on the PCL-M is suggestive of a PTSD diagnosis (Forbes, Creamer, & Biddle, 2001;Weathers et al., 1993). The PCL-M has been shown to have excellent internal consistency in veterans, victims of motor vehicle accidents, and sexual assault survivors as well as excellent test–retest reliability in veterans. In addition, the PCL has demon- strated excellent convergent validity with alternative measures of PTSD (Orsillo, 2002). The internal consistency in the present study was .87. TLFB.The TLFB is a retrospective measurement of daily substance use (Sobell & Sobell, 1992). The measure is com- pleted via a calendar format and at the direction of a trained assessor to enhance recall. In thepresent study, the TLFB as- sessed use of alcohol, stimulants (e.g., cocaine), opiates (e.g., heroin), marijuana, prescription drugs (e.g., prescription opioids, benzodiazepines, and psychostimulants), and nicotine over the past 60 days. The TLFB has demonstrated good psychometric proper- ties in the literature, including test–retest reliability, convergent and discriminant validity with other measures, and agreement with collateral informants and urine assays (Fals-Stewart, O’Farrell, Freitas, McFarlin, & Rutigliano, 2000). Data Analysis Of the initial 171 participants, 28 participants were excluded because of missing data on any of the primary measures. An additional 11 participants were excluded because of errors in their reporting on the TLFB. There were no differences between the excluded and included participants on demographics (ps .26), psychiatric diagnoses (ps .72), social support (p .08), PTSD symptoms (ps .26), or alcohol use (p .29). All remaining participants were included in the analyses. A series of hierarchical regression analyses were conducted to identify the unique relations among social support and PTSD symptoms and substance use. In the first step of each of these analyses, demographic variables (i.e., age, gender, race, relationship status, and employment status) were entered as covariates. Social support (DRRI) was entered in the second step. The analyses were conducted multiple times, with each measure of PTSD symptom severity (CAPS and PCL-M) and substance use (TLFB—Alcohol Use, TLFB—Stimulant Use, TLFB—Opiate Use, TLFB—Marijuana Use, TLFB—Prescription Drug Misuse) entered as a dependent variable. In addition, the nonmatching variable was included in the first step (PTSD as covariate for SUD as dependent; SUD as covariate for PTSD as dependent variable). The distribution of all dependent variableswas investigated to inform final inclusion of variables. Separate models were run with the CAPS and PCL-M to investigate the reliability of the findings across clinician-rated and self-reported PTSD symptoms. The CAPS and PCL are considered “gold stan- dards” in the measurement of PTSD and are frequently studied together in this way (Orsillo, 2002). Results Demographics of the Sample The average participant was 41.7 years old (SD 12.0), male (90.9%), White (50.0%) or Black (47.1%), and unemployed (68.9%). Most participants were either married (31.1%) or di- vorced/widowed (44.7%). The average number of years of educa- tion was 13.8 years (SD 1.8), and 59.1% had been deployed to OEF, OIF, or Operation New Dawn (OND). Most participants were diagnosed with PTSD (76.5%) and en- dorsed elevated symptoms consistent with a PTSD diagnosis on the PCL-M (M 59.3;SD 12.9). Most participants also were diagnosed with alcohol dependence (73.3%) and reported 25.0 total days used within the past 60 days (SD 21.2). Approxi- mately 38.9% also met criteria for current drug dependence. The average DRRI social support score was 46.6 (SD 8.8), and these scores were normally distributed within the sample (skewness 0.29; kurtosis 0.44). Regression Analyses Before completing the regression analyses, the distribution of each dependent variable was investigated. TLFB—Alcohol Use, CAPS, and PCL-M variables were found to have acceptable skew- ness (range 0.671 to 0.528) and kurtosis (range 0.674 to 0.376) and were included in the regression analyses. However, TLFB— Stimulant Use, TLFB—Opiate Use, TLFB—Marijuana Use, and TLFB—Prescription Drug Misuse evidenced unacceptable skew- ness (range 2.560 –7.746) and kurtosis (range 5.631–67.361). There were significant missing data in these variables that may have contributed to their skewness and kurtosis and greatly limited their possible interpretation. Thus, these variables were excluded from the analyses. In the analyses with PTSD as the dependent variable (seeTable 1), social support was used to predict the scales assessing PTSD symptoms with separate models for the CAPS and PCL-M scores. The first step included demographics and TLFB—Alcohol Use. The first steps in both models were significant (Fs 2.17;ps 0.05). The second step added social support and significantly increased the variance explained in both models (F change s 7.8; ps .007). With the addition of the second step, social support emerged with significant relation to PTSD symptoms in each of the models (ts 2.7;ps .004) above and beyond alcohol use, whereas greater social support was predictive of less severe PTSD symptoms in the CAPS and PCL-M. In the analyses with alcohol use as the dependent variable (see Table 1), demographic variables, PTSD symptoms (separate mod- els for CAPS and PCL-M symptoms), and social support were used to predict alcohol use. The first step with demographics and PTSD symptoms was significant in the model with the CAPS (F 2.72;p .02), but not significant in the model with the PCL-M 766 GROS ET AL. (F 1.72;p .122). The second step with social support was significant in both models (Fs 2.92;ps .008), with significant R 2s(F change 9.4;ps .004) as well as the social support variable (ts 3.0;p .004), whereas greater social support was predictive of greater alcohol use symptoms. Discussion The present study investigated the relations among PTSD, co- occurring SUD, and social support in veterans. Consistent with our hypotheses, social support had a significant relation to PTSD symptoms as well as alcohol use. Social support was associated with PTSD and alcohol use above and beyond the comorbid condition (self-reported PTSD as covariate for SUD as dependent; self-reported SUD as covariate for PTSD as dependent variable), suggesting that social support had an independent relation with each cluster of symptoms in this sample. Increased social support was associated with less severe PTSD symptoms. However, the alcohol use findings were in the opposite direction, with increased social support associated with increased alcohol use. Interestingly, average social support in this sample appears lower than averages reported in two recent studies also using the DRRI, including a VA treatment-seeking sample of OEF/OIF veterans (Pietrzak et al.,2010) as well as National Guard soldiers returning from OIF with or without new-onset PTSD (Polusny et al., 2011). Together, these findings highlight the significance of social support in veterans with symptoms of PTSD and alcohol use. The PTSD findings are consistent with previous findings for social support. Social support is associated with the lack of devel- opment of PTSD after trauma exposure (Kilpatrick et al., 2007; Pietrzak et al., 2010;Wilcox, 2010) and plays a significant role in the successful treatment of PTSD (Price et al., 2013). The current study is the first to investigate this relation in veterans with comorbid PTSD and SUD. Despite the noted relation between increased social support and increased alcohol use, increased so- cial support was associated with less severe PTSD symptoms. This finding is surprising because of the more severe symptoms when both disorders are present. That is, more severe PTSD symptoms are associated with increased SUD symptoms (McCauley et al., 2012). This relation among the three variables may further high- light the detrimental (decreased) nature of poor social support in PTSD independent of the presence of a SUD. The current findings regarding the relation between alcohol use and social support were in contrast to our hypothesis as well as the previous literature (Ren, Skinner, Lee, & Kazis, 1999;Sacco, Table 1 Hierarchical Regression Analysis Testing Social Support Predicting PTSD Symptoms and Alcohol Use Step VariableBSE tF R 2 Clinician Administered PTSD Scale (CAPS) 1 Age 0.41 0.19 .20 2.16 3.08 .128 Gender 7.54 7.11 .90 1.06 Race 3.41 2.60 .11 1.31 Relationship Status 4.02 2.60 .13 1.55 Employment Status 2.72 4.67 .05 0.58 TLFB—Alcohol Use 0.23 0.09 .22 2.55 2 DRRI—Social Support 0.91 0.24 .32 3.79 4.97 .090 PTSD Checklist–Military (PCL-M) 1 Age 0.13 0.10 .12 1.28 2.18 .092 Gender 5.23 3.80 .12 1.38 Race 0.16 1.27 .01 0.12 Relationship Status 2.95 1.40 .19 2.11 Employment Status 0.71 2.50 .03 0.28 TLFB—Alcohol Use 0.07 0.05 .13 1.46 2 DRRI—Social Support 0.37 0.13 .24 2.80 3.09 .052 TLFB–Total Days Used–Alcohol (with CAPS) 1 Age 0.08 0.19 .04 0.41 2.72 .115 Gender 9.57 6.77 .12 1.42 Race 2.88 2.48 .10 1.16 Relationship Status 2.44 2.50 .09 1.00 Employment Status 7.14 4.42 .14 1.62 CAPS—PTSD 0.21 0.08 .22 2.55 2 DRRI—Social Support 0.83 0.23 .31 3.57 4.37 .082 TLFB–Total Days Used–Alcohol (with PCL-M) 1 Age 0.05 0.18 .03 0.29 1.72 .074 Gender 7.91 6.80 .10 1.16 Race 0.66 2.27 .03 0.29 Relationship Status 0.49 2.54 .19 0.19 Employment Status 9.51 4.37 .19 2.18 PCL—PTSD 0.23 0.16 .13 1.46 2 DRRI—Social Support 0.71 0.23 .27 3.08 2.93 .064 Note.TLFB time line follow-back; DRRI Deployment Risk and Resiliency Inventory. p .05. p .01. p .001. 767 PAIN, PTSD, AND SUBSTANCE USE Bucholz, & Harrington, 2014). More specifically, previous re- search shows that although adaptive social support is consistently associated with successful substance use treatment outcomes, some studies have reported no significant relation between alcohol use severity and social support (Ren et al., 1999;Sacco et al., 2014) and others reporting a small negative relation between the two (Boscarino, 1995). The present findings suggest that increased social support was associated with increased alcohol use. Although the findings are contrary to the literature, there are a few possible hypotheses for these findings that may inform future investigation. First, the present study was completed on a much younger sample of veterans from recent combat theaters (e.g., OEF/OIF/OND), suggesting possible differences in veterans from varying eras of service. In addition, previous research has demonstrated robust positive associations between younger age and greater alcohol use in adult samples (Centers for Disease Control and Prevention, 2012) and that younger adults are more likely to drink in social settings to enhance social enjoyment (Gruenewald, Remer, & LaScala, 2014;O’Hara, Armeli, & Tennen, 2015). One recent study of heavy-drinking OEF/OIF veterans indicated that those with comorbid PTSD were more likely to attribute their alcohol misuse to symptom self-medication whereas those without PTSD were more likely to drink to enhance social experiences (McDevitt-Murphy, Fields, Monahan, & Bracken, 2015). Perhaps the younger veterans enrolled in our study were engaged in more socially rewarding drinking activities or had less opportunity to experience the negative consequences of prolonged heavy drink- ing, which may account for the perceived association between social support and alcohol use observed here. Finally, the literature examining associations between SUD and social support among veterans is limited, suggesting that differences in drinking patterns and associations with social support may have transitioned over time and with important contextual changes such as prolonged U.S. engagement in the conflicts in Iraq and Afghanistan. A further complicating factor is that social support is conceptualized and measured in widely varying ways across the literature. Additional research on similar samples attending to the nuances of social support source and type is needed to replicate the present findings and investigate these hypotheses among more current and repre- sentative veteran samples. Despite the interaction of PTSD and alcohol use symptoms, the overall level of social support was low in the present study and particularly when compared with other similar studies of veterans and with the same social support measure (Pietrzak et al., 2010; Polusny et al., 2011). This finding may suggest that the relation between more severe PTSD symptoms and decreased social sup- port may be much stronger than the relation between increased alcohol use and increased social support, resulting in overall lower social support. Because social support has been found to be protective against the development of PTSD and important in the related treatment outcome in veterans with PTSD, treatments for PTSD and associated comor- bidities should incorporate a social support building component to improve symptoms and potentially reduce future relapse. One exam- ple is the use of Concurrent Treatment of PTSD and Substance Use Disorders Using Prolonged Exposure (COPE;Back et al., 2015), which contains instruction to complete “healthy activities that you have lost interest” in as part of the “in vivo” exposure exercises. Additional examples include dyadic interventions targeting co-occurring PTSD and SUD such as Significant-Other Enhanced Cognitive–Behavioral Therapy (McDevitt-Murphy, 2011) and Cou- ple Treatment for AUD and PTSD (Schumm, Monson, O’Farrell, Gustin, & Chard, 2015). Both treatments aim to enhance social support gained from partners to simultaneously reduce symptomatol- ogy and improve dyadic functioning. Although additional research is needed, the use of treatments that encourage improvements in social support could have added benefits in PTSD psychotherapeutic out- comes. The present study contains several limitations. First, only single measures for social support and alcohol use were used, limiting their reliability and comparison across self-report and clinician- rated assessments. In addition, the sample contained an insufficient number of participants endorsing use of other substances of abuse; therefore, the findings cannot generalize to use of other substances. Although most veterans were deployed to OEF/OIF/OND, a mi- nority of participants were veterans from other conflicts. Finally, the study was limited to a cross-sectional investigation and cannot inform treatment or changes over time. Each of these limitations should be addressed in future research on this topic. The present study investigated the relations among social sup- port, PTSD, and SUD symptoms in a sample of veterans. The findings supported previous investigations on PTSD and social support, but they demonstrated a different pattern for SUD and social support. Together, these findings highlight the significance of social support among individuals with PTSD and SUD, and they further emphasize its potential importance in the prevention and treatment of these challenging and commonly co-occurring disor- ders. References Back, S. E. (2010). Toward an improved model of treating co-occurring PTSD and substance use disorders.The American Journal of Psychiatry, 167,11–13.http://dx.doi.org/10.1176/appi.ajp.2009.09111602 Back, S. E., Foa, E. B., Killeen, T. K., Mills, K. L., Teesson, M., Dansky Cotton, B., & Brady, K. T. (2015).Concurrent treatment of PTSD and substance use disorders using prolonged exposure (COPE): Therapist guide. New York, NY: Oxford Press. Back, S. E., Killeen, T., Foa, E. B., Santa Ana, E. J., Gros, D. F., & Brady, K. T. (2012). Use of prolonged exposure to treat PTSD in an Iraq veteran with comorbid alcohol dependence.The American Journal of Psychia- try, 169,688 – 691.http://dx.doi.org/10.1176/appi.ajp.2011.11091433 Back, S. E., Waldrop, A. E., & Brady, K. T. (2009). Treatment challenges associated with comorbid substance use and posttraumatic stress disor- der: Clinicians’ perspectives.The American Journal on Addictions, 18, 15–20.http://dx.doi.org/10.1080/10550490802545141 Blake, D. D., Weathers, F. W., Nagy, L. M., Kaloupek, D. G., Gusman, F. D., Charney, D. S., & Keane, T. M. (1995). The development of a clinician-administered PTSD scale.Journal of Traumatic Stress, 8, 75–90.http://dx.doi.org/10.1002/jts.2490080106 Blanco, C., Xu, Y., Brady, K., Pérez-Fuentes, G., Okuda, M., & Wang, S. (2013). Comorbidity of posttraumatic stress disorder with alcohol de- pendence among US adults: Results from National Epidemiological Survey on Alcohol and Related Conditions.Drug and Alcohol Depen- dence, 132,630 – 638.http://dx.doi.org/10.1016/j.drugalcdep.2013.04 .016 Boscarino, J. A. (1995). Post-traumatic stress and associated disorders among Vietnam veterans: The significance of combat exposure and social support.Journal of Traumatic Stress, 8,317–336.http://dx.doi .org/10.1002/jts.2490080211 768 GROS ET AL. Bowe, A., & Rosenheck, R. (2015). PTSD and substance use disorder among veterans: Characteristics, service utilization and pharmacother- apy.Journal of Dual Diagnosis, 11,22–32.http://dx.doi.org/10.1080/ 15504263.2014.989653 Brancu, M., Thompson, N. L., Beckham, J. C., Green, K. T., Calhoun, P. S., Elbogen, E. B.,…theVAMid-Atlantic MIRECC Registry Workgroup. (2014). The impact of social support on psychological distress for U.S. Afghanistan/Iraq era veterans with PTSD and other psychiatric diagnoses.Psychiatry Research, 217,86 –92.http://dx.doi .org/10.1016/j.psychres.2014.02.025 Brewin, C. R., Andrews, B., & Valentine, J. D. (2000). Meta-analysis of risk factors for posttraumatic stress disorder in trauma-exposed adults. Journal of Consulting and Clinical Psychology, 68,748 –766.http://dx .doi.org/10.1037/0022-006X.68.5.748 Campbell, D. G., Felker, B. L., Liu, C. F., Yano, E. M., Kirchner, J. E., Chan, D.,…Chaney, E. F. (2007). Prevalence of depression-PTSD comorbidity: Implications for clinical practice guidelines and primary care-based interventions.Journal of General Internal Medicine, 22, 711–718.http://dx.doi.org/10.1007/s11606-006-0101-4 Centers for Disease Control and Prevention. (2012). Vital signs: Binge drinking prevalence, frequency, and intensity among adults—United States, 2010.MMWR. Morbidity and Mortality Weekly Report, 61, 14 –19. Cohen, L. R., & Hien, D. A. (2006). Treatment outcomes for women with substance abuse and PTSD who have experienced complex trauma. Psychiatric Services, 57,100 –106.http://dx.doi.org/10.1176/appi.ps.57 .1.100 DeBeer, B. B., Kimbrel, N. A., Meyer, E. C., Gulliver, S. B., & Morissette, S. B. (2014). Combined PTSD and depressive symptoms interact with post-deployment social support to predict suicidal ideation in Operation Enduring Freedom and Operation Iraqi Freedom veterans.Psychiatry Research, 216,357–362.http://dx.doi.org/10.1016/j.psychres.2014.02 .010 Fals-Stewart, W., O’Farrell, T. J., Freitas, T. T., McFarlin, S. K., & Rutigliano, P. (2000). The Timeline Followback reports of psychoactive substance use by drug-abusing patients: Psychometric properties.Jour- nal of Consulting and Clinical Psychology, 68,134 –144.http://dx.doi .org/10.1037/0022-006X.68.1.134 Forbes, D., Creamer, M., & Biddle, D. (2001). The validity of the PTSD checklist as a measure of symptomatic change in combat-related PTSD. Behaviour Research and Therapy, 39,977–986.http://dx.doi.org/10 .1016/S0005-7967(00)00084-X Gruenewald, P. J., Remer, L. G., & LaScala, E. A. (2014). Testing a social ecological model of alcohol use: The California 50-city study.Addiction, 109,736 –745.http://dx.doi.org/10.1111/add.12438 Kaier, E., Possemato, K., Lantinga, L. J., Maisto, S. A., & Ouimette, P. C. (2014). Associations between PTSD and healthcare utilization among OEF/OIF veterans with hazardous alcohol use.Traumatology, 20,142– 149.http://dx.doi.org/10.1037/h0099399 Kilpatrick, D. G., Koenen, K. C., Ruggiero, K. J., Acierno, R., Galea, S., Resnick, H. S.,…Gelernter, J. (2007). The serotonin transporter genotype and social support and moderation of posttraumatic stress disorder and depression in hurricane-exposed adults.The American Journal of Psychiatry, 164,1693–1699.http://dx.doi.org/10.1176/appi .ajp.2007.06122007 King, L. A., King, D. W., Vogt, D. S., Knight, J., & Samper, R. E. (2006). Deployment Risk and Resilience Inventory: A collection of measures for studying deployment-related experiences of military personnel and vet- erans.Military Psychology, 18,89 –120.http://dx.doi.org/10.1207/ s15327876mp1802_1 Laffaye, C., Cavella, S., Drescher, K., & Rosen, C. (2008). Relationships among PTSD symptoms, social support, and support source in veterans with chronic PTSD.Journal of Traumatic Stress, 21,394 – 401.http:// dx.doi.org/10.1002/jts.20348Manuel, J. K., McCrady, B. S., Epstein, E. E., Cook, S., & Tonigan, J. S. (2007). The pretreatment social networks of women with alcohol de- pendence.Journal of Studies on Alcohol and Drugs, 68,871– 878. http://dx.doi.org/10.15288/jsad.2007.68.871 McCauley, J. L., Killeen, T., Gros, D. F., Brady, K. T., & Back, S. E. (2012). Posttraumatic stress disorder and co-occurring substance use disorders: Advances in assessment and treatment.Clinical Psychology: Science and Practice, 19,283–304.http://dx.doi.org/10.1111/cpsp .12006 McCrady, B. S. (2004). To have but one true friend: Implications for practice of research on alcohol use disorders and social network.Psy- chology of Addictive Behaviors, 18,113–121.http://dx.doi.org/10.1037/ 0893-164X.18.2.113 McDevitt-Murphy, M. E. (2011). Significant other enhanced cognitive- behavioral therapy for PTSD and alcohol misuse in OEF/OIF veterans. Professional Psychology: Research and Practice, 42,40 – 46.http://dx .doi.org/10.1037/a0022346 McDevitt-Murphy, M. E., Fields, J. A., Monahan, C. J., & Bracken, K. L. (2015). Drinking motives among heavy-drinking veterans with and without posttraumatic stress disorder.Addiction Research & Theory, 23, 148 –155.http://dx.doi.org/10.3109/16066359.2014.949696 O’Hara, R. E., Armeli, S., & Tennen, H. (2015). College students’ drinking motives and social-contextual factors: Comparing associations across levels of analysis.Psychology of Addictive Behaviors, 29,420 – 429. http://dx.doi.org/10.1037/adb0000046 Orsillo, S. (2002). Measures for acute stress disorder and posttraumatic stress disorder. In M. M. Antony, S. Orsillo, & L. Roemer (Eds.), Practitioner’s guide to empirically based measures of anxiety(pp. 255–307). New York, NY: Kluver Publications.http://dx.doi.org/10 .1007/0-306-47628-2_20 Petrakis, I. L., Rosenheck, R., & Desai, R. (2011). Substance use comor- bidity among veterans with posttraumatic stress disorder and other psychiatric illness.The American Journal on Addictions, 20,185–189. http://dx.doi.org/10.1111/j.1521-0391.2011.00126.x Pietrzak, R. H., Goldstein, R. B., Southwick, S. M., & Grant, B. F. (2011). Prevalence and Axis I comorbidity of full and partial posttraumatic stress disorder in the United States: Results from Wave 2 of the National Epidemiologic Survey on Alcohol and Related Conditions.Journal of Anxiety Disorders, 25,456 – 465.http://dx.doi.org/10.1016/j.janxdis .2010.11.010 Pietrzak, R. H., Johnson, D. C., Goldstein, M. B., Malley, J. C., Rivers, A. J., Morgan, C. A., & Southwick, S. M. (2010). Psychosocial buffers of traumatic stress, depressive symptoms, and psychosocial difficulties in veterans of Operations Enduring Freedom and Iraqi Freedom: The role of resilience, unit support, and postdeployment social support. Journal of Affective Disorders, 120,188 –192.http://dx.doi.org/10.1016/ j.jad.2009.04.015 Pietrzak, R. H., Johnson, D. C., Goldstein, M. B., Malley, J. C., & Southwick, S. M. (2009). Psychological resilience and postdeployment social support protect against traumatic stress and depressive symptoms in soldiers returning from Operations Enduring Freedom and Iraqi Freedom.Depression and Anxiety, 26,745–751. Polusny, M. A., Erbes, C. R., Murdoch, M., Arbisi, P. A., Thuras, P., & Rath, M. B. (2011). Prospective risk factors for new-onset post- traumatic stress disorder in National Guard soldiers deployed to Iraq. Psychological Medicine, 41,687– 698.http://dx.doi.org/10.1017/ S0033291710002047 Price, M., Gros, D. F., Strachan, M., Ruggiero, K. J., & Acierno, R. (2013). The role of social support in exposure therapy for Operation Iraqi Freedom/Operation Enduring Freedom veterans.: A preliminary inves- tigation.Psychological Trauma: Theory, Research, Practice, and Pol- icy, 5,93–100.http://dx.doi.org/10.1037/a0026244 Ren, X. S., Skinner, K., Lee, A., & Kazis, L. (1999). Social support, social selection and self-assessed health status: Results from the veterans 769 PAIN, PTSD, AND SUBSTANCE USE health study in the United States.Social Science & Medicine, 48, 1721–1734.http://dx.doi.org/10.1016/S0277-9536(99)00069-6 Sacco, P., Bucholz, K. K., & Harrington, D. (2014). Gender differences in stressful life events, social support, perceived stress, and alcohol use among older adults: Results from a national survey.Substance Use & Misuse, 49,456 – 465.http://dx.doi.org/10.3109/10826084.2013.846379 Schumm, J. A., Monson, C. M., O’Farrell, T. J., Gustin, N. G., & Chard, K. M. (2015). Couple treatment for alcohol use disorder and posttrau- matic stress disorder: Pilot results from US military veterans and their partners.Journal of Traumatic Stress, 28,247–252.http://dx.doi.org/10 .1002/jts.22007 Seal, K. H., Bertenthal, D., Miner, C. R., Sen, S., & Marmar, C. (2007). Bringing the war back home: Mental health disorders among 103,788 US veterans returning from Iraq and Afghanistan seen at Department of Veterans Affairs facilities.Archives of Internal Medicine, 167,476 – 482.http://dx.doi.org/10.1001/archinte.167.5.476 Sheehan, D. V., Lecrubier, Y., Sheehan, K. H., Amorim, P., Janavs, J., Weiller, E.,…Dunbar, G. C. (1998). The Mini-International Neuro- psychiatric Interview (M. I. N. I.): The development and validation of a structured diagnostic psychiatric interview forDSM–IVand ICD-10. Journal of Clinical Psychiatry, 59(Suppl. 20), 22–33. Sobell, L. C., & Sobell, M. B. (1992). Timeline follow-back: A technique for assessing self-reported alcohol consumption. In R. Z. Litten & J. P. Allen (Eds.),Measuring alcohol consumption: Psychosocial and bio-chemical methods(pp. 41–72). Totowa, NJ: Humana Press.http://dx.doi .org/10.1007/978-1-4612-0357-5_3 Vogt, D. S., Proctor, S. P., King, D. W., King, L. A., & Vasterling, J. J. (2008). Validation of scales from the Deployment Risk and Resilience Inventory in a sample of Operation Iraqi Freedom veterans.Assessment, 15,391– 403. Weathers, F. W., Litz, B. T., Herman, D. S., Huska, J. A., & Keane, T. M. (1993).The PTSD Checklist: Reliability, validity and diagnostic utility. Presented at the annual meeting of the International Society for Trau- matic Stress Studies, San Antonio, TX. Whealin, J. M., DeCarvalho, L. T., & Vega, E. M. (2008).Strategies for managing stress after war: Veteran’s workbook and guide to wellness. Hoboken, NJ: Wiley. Wilcox, S. (2010). Social relationships and PTSD symptomatology in combat veterans.Psychological Trauma: Theory, Research, Practice, and Policy, 2,175–182. Zywiak, W. H., Longabaugh, R., & Wirtz, P. W. (2002). Decomposing the relationships between pretreatment social network characteristics and alcohol treatment outcome.Journal of Studies on Alcohol, 63,114 –121. Received March 3, 2016 Revision received July 5, 2016 Accepted July 7, 2016 770 GROS ET AL.
LRSC
MAY–JUNE 2012 47 Magdalena Kaczmarek is an associate professor at Warsaw School of Social Psychology, Faculty of Psychology. Bogdan Zawadzki is a professor at the Uni- versity of Warsaw, Faculty of Psychology. Address correspondence to Magdalena Kaczmarek, Warsaw School of Social Psychology, Chodakowska 19/31, 03-815 Warsaw, Poland; e-mail: [email protected] The research was supported by Grant PL0088 “Psychological Causes and Consequences of Traffic Accidents,” financed by the Financial Mechanism Committee established by Iceland, Liechtenstein, and Norway through the EEA Financial Mechanism and Polish Ministry of Sciences and Higher Education. 47 Journal of Russian and East European Psychology, vol. 50, no. 3, May–June 2012, pp. 47–64. © 2012 M.E. Sharpe, Inc. All rights reserved. Permissions: www.copyright.com ISSN 1061–0405 (print)/ISSN 1558–0415 (online) DOI: 10.2753/RPO1061-0405500303 M a g d a l e n a K a c z Ma r e K a n d B o g d a n z a w a d z Ki Exposure to Trauma, Emotional Reactivity, and Its Interaction as Predictors of the Intensity of PTSD Symptoms in the Aftermath of Motor Vehicle Accidents The aim of the study was to analyze the relationship between emotional reactivity, exposure to trauma and its interaction, and posttraumatic stress disorder (PTSD) symptoms in motor vehicle accident (MVA) survivors. Both emotional reactivity and exposure to trauma were expected to be significant predictors of the intensity of PTSD symp- toms. Exposure to trauma was considered as a total score but also as successive indexes, such as threat to life during an accident, injuries sustained in the aftermath of an accident, emotions felt during an ac- cident, dissociation experiences, and the amount of material losses. 48 JOURNAL OF RUSSIAN AND EAST EUROPEAN PSYCHOLOGY Emotional reactivity was also expected to play the role of moderator, increasing the positive relationship between exposure to trauma and PTSD symptoms in people characterized by higher emotional reactiv- ity. The analyses were performed in two separate samples. The first sample consists of 458 MVA survivors who had a traffic collision up to six months before the study. The second sample (n = 674) comprises MVA survivors who had an accident more than six but less than twenty- four months before the study. The correlation and regression analyses revealed that, as expected, both emotional reactivity and exposure to trauma are significant predictors of PTSD symptoms, which explained about one-third of total variance of symptoms. The hierarchical regres- sion with interactions between emotional reactivity and exposure to trauma also supports the hypothesis that emotional reactivity can be a moderator and the MVA survivors who are more emotionally reactive develop more intensive PTSD symptoms when confronted with severe and stressful experience. Research on trauma, and especially on disasters and accidents, and its psychological consequences such as posttraumatic stress disorder (PTSD) does not have a long history in Polish psychol- ogy. It has been developed mainly in the domain of domestic violence or—in psychiatry—as studies on World War II veterans and survivors of concentration camps (Lis-Turlejska, 1998). In 1997, a huge flood disaster occurred in the southern and western part of Poland, in which fifty-five people lost their lives and prop – erty damages were calculated at $3.5 billon. In the aftermath, the Polish Ministry of Science announced a research program titled “Man Under Disaster,” led by our research group. The aims of this grant were to study the consequences of such experience as well as the risk and protective factors that predict social and psychological functioning in survivors. Being interested in indi- vidual differences and personality processes of self-regulation, we included personality variables, such as temperament traits, personality traits from the “big five” model of personality, and coping styles (see Strelau and Zawadzki, 2005). Our studies on disaster were usually conducted using the family research model: we carried out longitudinal studies with two or three repeated measurements and maintained a focus on MAY–JUNE 2012 49 aspects of consequences such as material losses and financial difficulties. Since 1997, our research group conducted studies on several samples of flood victims, on a sample of coal miners who had experienced mining accidents, and on families who had experienced domestic fire accidents (Kaczmarek, KaŸ mierczak, and Strelau, 2009). In 2008 we began a research program on motor vehicle ac- cident (MVA) survivors, which is accompanied by a therapeutic program addressing people who suffer PTSD in the aftermath of an MVA. In this program we studied individual victims as well as a small group of families, although the results presented here include only the group of individual victims without their family members. Posttraumatic stress disorder is a set of symptoms described in DSM IV (APA, 1994) and defined as a possible sequel of experience of a traumatic event. The symptoms of PTSD are defined using three diagnostic criteria, each composed of similar kinds of symptoms. In short, PTSD is manifested as: (1) symptoms of reexperiencing the event, such as intrusive thoughts or dreams; (2) symptoms of numbness, expressing emotional coldness or decreasing activity, and in symptoms of avoiding stimuli associated with the traumatic event; and (3) symptoms of the state of hyperarousal, for example, as expressed in difficulties concentrating or falling asleep. Focusing on PTSD as the specific consequence of traumatic experience, in our research we take into account the whole diversity of severity of symptoms, so we are interested in the intensity of PTSD symptoms as a dimension and not only in PTSD as a syndrome in a strictly clinical meaning. In this approach, PTSD symptoms can be con- sidered as a measure of distress in the aftermath of exposure to an extreme stressor. The empirical support of the validity of such an approach is provided by the analyses conducted by Ruscio, Ruscio, and Keane (2002). As mentioned above, we are interested in personality, includ- ing temperament factors, as predictors of the intensity of PTSD symptoms. Temperament, defined as personality traits that are present since early childhood, can be observed not only in human 50 JOURNAL OF RUSSIAN AND EAST EUROPEAN PSYCHOLOGY behavior but also in animals, and refers to formal aspects of be- havior (Strelau, 1998). Formal characteristics of behavior can be considered in terms of energetic and temporal patterns of behav- ior. Temperament traits, being more or less unspecific, penetrate all kinds of behavior, whatever the content or direction of this behavior. Following the Regulative Theory of Temperament, for – mulated by Strelau and described in detail in several publications (Strelau, 1998, 2008), temperament plays a regulative role that consists of modifying (moderating) the stimulative and temporal value of situations and behavior according to individual-specific temperament traits. This role is especially evident in difficult situations and extreme behaviors. A situation of extreme stress that is highly demanding and puts pressure on individual abilities to cope is an example of such a difficult situation. Figure 1 displays the model of relationships between tempera- ment and stress phenomena: a system of complex and reciprocal relationships. Besides the role of temperament in moderating the intensity of the state of stress, studies indicate some further relationships within the model. Temperament (1) codetermines the intensity of stressors and, in the case of stressors dependent on the individual, also their probability of occurrence, (2) moderates coping efforts, and (3), contributes to the psychophysiological and/or psychological costs of the state of stress (Strelau, 2008). The Regulative Theory of Temperament postulates six traits: briskness, perseveration (which are related to the temporal level of behavior), emotional reactivity, endurance, sensory sensitiv- ity, and activity (which describe the energetic level of behavior). All of these can play an important role as individual predictors of functioning under stress, but as the results of many studies show, emotional reactivity, which is defined as tendency to react intensively to emotion-generating stimuli, expressed in high emotional sensitivity and in low emotional endurance, seems to be highly represented and the most consistent among different samples related to the consequences of stress (cf. Strelau, 2008). In the extreme stress domain, where severe stressors that are in- dependent of the individual are taken into consideration, the role MAY–JUNE 2012 51 of temperament seems to be similar. In flood survivors, it has been established that emotional reactivity (ER) is the temperamental trait that consistently explains, in different age samples, the major part of the intensity of PTSD symptoms (Strelau, Kaczmarek, and Zawadzki, 2006). In MVA survivors, it is also expected that ER should be related to PTSD symptoms and ER should explain the significant amount of variance in PTSD symptoms (Hypothesis 1). However, it is also expected that ER may play a role as a moderator of individual reactions to the stressor—the subjects who are more emotionally reactive should react to an accident with more intensive symptoms of PTSD when confronted with a more severe and stressful experience (Hypothesis 2). St ress ors inde pend ent on the individ ua l • Pa ren t-ch ild conflic t • Ma rita l di scor d • Challenges /demands St ress ors inde pend ent of th e ind ivid ua l • Na tura l di sa ster s • Deat h of cl os e pers on Temp erament Copi ng St ate of stress Consequences of stress • Be havior dist urbanc es • Heal th ch ange s • Psyc hologic al ch anges • Ph ysiologic al and bioc hemi cal ch anges • Man-made disasters Figure 1. Model of Relationships Between Temperament and Stress Phenomena Source: J. Strelau, Temperament as a Regulator of Behavior (New York: Eliot Werner, 2008), p. 121. 52 JOURNAL OF RUSSIAN AND EAST EUROPEAN PSYCHOLOGY Besides individual factors, such as temperament traits, many pre-, peri-, and posttraumatic factors are important predictors of adaptation after extreme stress. The meta-analyses in this field have mainly revealed the significant role of peri- and posttrau- matic variables, especially those related to the level of exposure to trauma, reactions of an individual to this experience, and its further consequences, including secondary stressors (Brewin, Andrews, and Valentine, 2000; Ozer et al., 2003). In terms of the definition of PTSD (APA, 1994), trauma involves a threat to life or health experience as well as the accompanying emotions of intense fear, hopelessness, and horror. The more life and health was endangered and the more physically and emotionally harmful the experience was, the more difficult the process of recovering is. This relationship—between exposure to trauma and intensity of PTSD symptoms—has been established in almost every study in this area. Going beyond the definition of trauma, two other factors describing the stressful experience seem to be important: dissociations during an accident and the amount of material losses in the aftermath of an accident. In meta-analyses conducted by Ozer and colleagues (2003), peritraumatic dissociations were listed among the most important predictors of PTSD symptoms. Dissociations include a mental blank regarding the accident as well as a sense of being detached from emotions and experiencing an accident as something un- real, not happening in reality. On the other hand, Brewin and her coworkers (Brewin, Andrews, and Valentine, 2000) note the role of additional life stress in predicting PTSD. Material losses and the ensuing financial troubles may be considered an example of additional life stress, and they occur in all kinds of traumatic experience in which property damages occur. In conclusion, exposure to trauma (ET), understood as a threat to life and health and peritraumatic emotions, as well as peri- traumatic dissociation and the amount of material losses, should be related to the intensity of PTSD symptoms and ET should explain a significant amount of variance of PTSD symptoms (Hypothesis 3). MAY–JUNE 2012 53 Method Subjects The analysis was conducted using two samples of MVA survivors: sample A1 consists of 458 participants who had an MVA up to six months before the data were gathered, and sample A2 is composed of 674 participants who had traffic accident six to twenty-four months before. Both samples were surveyed once, which constitutes the first wave in our longitudinal study. All of the subjects are adults who were directly involved in an MVA as drivers, passengers, or pedestrians. The most frequent type of MVA was a crash between cars or between a car and a motorbike. In both samples, men are in the majority, which is in accordance with data on the entire population of MVA victims in Poland. Almost all of the subjects are from Warsaw or places bordering Warsaw. Accidents took place in 2008–9. The details are presented in Table 1. Measures The PTSD-Factorial Version (PTSD-F), an inventory developed by Strelau and coworkers (Strelau et al., 2002), was applied for measuring the intensity of posttraumatic stress disorder symptoms. The PTSD-F is designed to assess the intensity of PTSD symptoms along the whole dimension describing the whole range of intensity of PTSD symptoms observed in flood victims. Apart from a total score, PTSD-F comprises two scales corresponding to two basic PTSD factors—intrusion/hyperarousal and avoidance/numbing. However, in further analyses only the total score, which was calculated as the sum of answers to all thirty items, scoring from zero to three points each, was taken into account. The total score reflects the general intensity of PTSD symptoms (in the analysis presented here, scores were corrected for gender and age via linear regression and saved as standardized residuals). Temperament was assessed in both samples by the Formal Characteristics of Behavior–Temperament Inventory (FCB-TI; 54 JOURNAL OF RUSSIAN AND EAST EUROPEAN PSYCHOLOGY Table 1 Demographic Characteristics of the Investigated Samples Sample Time after motor vehicle accident (MVA) NGender Age M (SD) Age range A1 1–6 months (M = 3.65, SD = 1.86) 458Female 200 Male 258 34.42 (13.45)18–66 A2 7–24 months (M = 13.18, SD = 3.96) 674Female 264 Male 410 37.06 (13.56)18–66 Type of MVA Education A1 Collision—cars or motorbikes, 73.0%; driving off the road, 10.9%; hitting a pedestrian, 11.1%; or a biker, 5.0% Lower, 10.9%; higher, 88.9%; lack of data, .2%; mostly college, 52.4% A2 Collision—cars or motorbikes, 69.7%; driving off the road, 12.3%; hitting a pedestrian, 12.9%; or a biker, 5.1% Lower, 17.1%; higher, 82.6%; lack of data, .3%; mostly college, 45.4% Notes: Education was coded as “lower” (1) elementary or trade school; and “higher” (2) for college or university level. Gender was coded: female (1) and male (2). Samples differed with regard to age (t = 3.21, df = 1130, p < .01) and education (χ 2 = 8.30, df = 1, p < .01), but not to gender (χ 2 = 2.28, df = 1, n.s.) and type of MVA (χ 2 = 1.54, df = 3, n.s.). MAY–JUNE 2012 55 Strelau and Zawadzki, 1993, 1995) which is composed of six scales: briskness, perseveration, sensory sensitivity, endurance, emotional reactivity, and activity. All scales contain twenty items each scored in “yes–no” format (the scores range from zero to twenty points). For the purpose of the analyses, only scores of the emotional reactivity scale were taken into account (as in the case of PTSD, scores of the ER scale were corrected for gender and age via linear regression and saved as standardized residuals). Exposure to trauma was measured by several questions on a short self-assessed survey developed for this study. The level of threat to life was measured by three questions about being in danger in reference to the participant and to the other people involved in the accident. The level of physical harm was also assessed by three questions about injuries to the body and their severity in reference to the subject and to the other people involved in the accident. Peritraumatic emotions were assessed by three questions about fear, hopelessness, and horror experienced dur – ing the trauma, and peritraumatic dissociation by five questions addressing the experience, such as a subjective sense of numbing, reduced awareness, derealization, depersonalization, and amnesia during or just after the MVA. The amount of material losses was assessed by only one question about a general appraisal of such losses in an accident. All of these variables were summed into appropriate total scores: life threat, injuries, peritraumatic emo- tions, dissociation, and material losses, and corrected for gender and age via linear regression. The total index was obtained by summing up all five variables, which can be considered as the measure of exposure to trauma. Procedure The personal data of MVA survivors were retrieved from the data- base of traffic incidents maintained by the Polish National Police Headquarters. First, all subjects were sent letters containing infor – mation about the research and offers of therapy for those who need it. The participants were surveyed in their homes by professional pollsters. All of them were asked to fill out a set of self-assessed 56 JOURNAL OF RUSSIAN AND EAST EUROPEAN PSYCHOLOGY questionnaires. Subjects were paid for participating in study. The procedure was approved by the local Science Ethics Committee. Results Analyses were conducted separately for each sample. A model based on correlations and hierarchical regressions was applied. The analyses began with calculating the correlation coefficients between the intensity of PTSD symptoms and separate indexes of exposure to trauma and emotional reactivity. The findings are shown in Table 2 in the first and second columns. The correlation coefficients indicate a moderate but significant pattern of relationships between PTSD symptoms and each of the indexes of exposure to trauma as well as emotional reactivity. All of these findings support hypotheses (H1 and H3). The lowest cor – relation was revealed in reference to threat to life, which can be considered a rather unexpected finding. 1 Comparing the samples reveals a quite similar pattern of correlation, which means that the time elapsed since the accident does not influence relationships between PTSD symptoms and individual measures of exposure to trauma and emotional reactivity. The one exception is peritraumatic dissociation, which is slightly more strongly related to PTSD symp- toms in the second sample. The exposure to trauma calculated as a total score is related to PTSD symptoms at a level of .49 (sample A1) or .55 (sample A2), whereas emotional reactivity correlates with PTSD symptoms of .39 (sample A1) or .38 (sample A2). Both exposure to trauma (ET) and emotional reactivity (ER) are posi- tively and significantly related to PTSD symptoms. All of these cor – relations are in line with data from our previous studies performed on several samples of flood survivors—the average correlations of ER and PTSD symptom intensity for comparable periods of time were equal to .35 and .36, respectively (see Zawadzki, Kaczmarek, and Strelau, 2009). However, in the present study slightly higher correlations were revealed in reference to exposure to trauma—the difference was significant (t = 1.96, df = 455; and t = 5.48, df = 671, p < .05; see Cohen and Cohen, 1983). This pattern of rela- tionships differs from our previous results obtained on samples of MAY–JUNE 2012 57 Table 2 Indicators of Exposure to Trauma, Emotional Reactivity, and Intensity of PTSD Symptoms Analysis Simple correlationsRegresssion (semipartial correlations) Regresssion (semipartial correlations) Indicator Sample A1Sample A2Sample A1Sample A2Sample A1Sample A2 Threat to life .21*.25* ———— Injuries to the body .35*.39*.17*.20*.16*.21* Peritraumatic emotions .39*.40*.21*.18*.17*.13* Peritraumatic dissociation .33*.41*.12*.20*.10*.17* Amount of material loses .35*.36*.19*.17*.19*.16* Emotional reactivity .39*.38* xx.29* .27* R (R 2) xx.52* (.27) .57* (.32).60* (.36).63* (.39) Exposure to trauma .49*.55* xx.43* .48* Emotional reactivity .39*.38* xx.31* .28* R (R 2) xxxx.58* (.33) .61* (.37) Note: The stepwise procedure was applied in regression analysis for all variables, using a hierarchical model, with emotional reactivity entered by the stepwise method in the second step. Sample A1: F = 42.55, df = 4/453, p < .01 and F = 49.69, df = 5/452, p < .01, F change = 57.17, df = 1/452, p < .01. Sample A2: F = 79.09, df = 4/669, p < .01 and F = 86.47, df = 5/668, p < .01, F change = 79.06, df = 1/668, p < .01. For emotional reactivity and trauma exposure—Sample A1: F = 112.69, df = 2/455, p < .01; Sample A2: F = 201.58, df = 2/671, p < .01; *correlation coef- ficient significant at p < .05; (—) variable removed from the model by stepwise method; x: variable not taken into account in the model. All analyses were done on standardized scores, corrected for age and gender. R —multiple correlation; R 2—explained variance. 58 JOURNAL OF RUSSIAN AND EAST EUROPEAN PSYCHOLOGY flood survivors, for whom the stronger correlations were recorded for ER than for trauma indexes, probably due to the more severe trauma exposure (threat to life, injuries, etc.) in the case of MVAs compared to natural disasters. Regression analyses were performed. In Table 2, the third and the fourth columns show the results of one-step analysis, in which only indexes of exposure to trauma were introduced into the model. In both samples this model explains about 30 percent of variance of PTSD symptoms. The two-step analyses, putting emotional reactivity in the second step, are presented in the last two columns of Table 2. The findings show that adding the temperament traits allows us to explain another portion of variance. In sample A1, ER adds 9 percent of the explained variance of PTSD symptoms, and in sample A2, 7 percent. Exposure to trauma and emotional reactivity in both samples seem to be significant predictors of PTSD symptoms. Exposure to trauma calculated as a total score is the strongest predictor of PTSD and, once again, slightly stronger than emotional reactivity. Finally, three-step regression analyses were done. In the last step, interactions, calculated as products between exposure to trauma measures and emotional reactivity, were introduced into the models. The results are presented in Table 3. Table 3 shows only the last step of these hierarchical analyses. All interactions are significant but weak predictors of PTSD symptoms. The results of this analysis enable us to consider the more general index of exposure to trauma and its interaction with emotional reactivity. 2 The bottom part of Table 3 shows the results of the last step of hierarchical regression analysis, in which emotional reactivity, total score of exposure to trauma, and its interactions in both samples were introduced into the model. All of these are significant predictors of PTSD symptoms. These interactions are shown in Figure 2. Although weak, the interactions show that the intensity of PTSD symptoms was highest when extremely emotional-reactive people were confronted with more threatening circumstances of an accident. This supports hypothesis 2, which expected emotional reactivity to play a role as moderator of reaction to trauma. As in MAY–JUNE 2012 59 the case of particular indexes of trauma exposure, interaction of the total index with ER was a significant but weak predictor of PTSD symptoms, and added 4 percent of explained variance of PTSD symptoms in sample A1, and 3 percent in sample A2. Discussion Analyses of the data showed that all of the formulated hypotheses were supported. Both emotional reactivity and exposure to trauma indexes are positive and significant predictors of the intensity of Table 3 Intensity of Symptoms of PTSD and Interactions Between Emotional Reactivity and Indexes of Exposure to Trauma as Well as the Final Model—Exposure to Trauma and Emotional Reactivity (semipartial correlations) Regression analysis Two-variable model (enter method) Variables and two-way interactions Sample A1Sample A2 Emotional reactivity × threat to life .14*.10* Emotional reactivity × injuries of the body .10*.10* Emotional reactivity × amount of material loses .15*.08* Emotional reactivity × peritraumatic emotions .11*.07* Emotional reactivity × peritraumatic dissociation .16*.14* Exposure to trauma .43*.50* Emotional reactivity .30*.28* Emotional reactivity × exposure to trauma .19*.17* Notes: *Correlation coefficient significant at p < .05; (–) variables removed from the model by stepwise method. In the model for two variables with their interaction, the enter method was applied. All analyses were done on standardized scores, corrected for age and gender (interactions were not standardized; see L.S. Aiken and S.G. West, Multiple Regression: Testing and Interpreting Interactions [Newbury Park, CA: Sage], 1991). For emotional reactivity and trauma exposure—Sample A1: F = 87.55, df = 3/454, p < .01; R = .61, R 2 = .37 (for interaction: DR 2 = 0.04, F change = 25.26, df = 1/454, p < .01); Sample A2: F = 151.45, df = 3/670, p < .01; R = .64, R2 = .40 (for interaction: DR 2 = 0.03, F change = 32.36, df = 1/670, p < .01). 60 JOURNAL OF RUSSIAN AND EAST EUROPEAN PSYCHOLOGY PTSD symptoms in motor vehicle accident survivors. It holds true in reference to both samples—survivors up to six months after the accident, and accident survivors who were involved in a collision or other traffic accident six to twenty-four months before the study. It was also revealed that interactions between emotional reactivity and trauma exposure (as total score and successive indexes) were positive, weak but significant predictors of PTSD symptoms. In summary, the results show that in the MVA survivors who were studied, both emotional reactivity as a temperament trait and expo- sure to trauma indexes are significant predictors of maladaptation in the aftermath of an accident, but exposure to trauma seems to be a slightly stronger predictor. All of the factors taken into ac- count explained about 33 percent of variance of PTSD symptoms, the same value reached in our previous studies of flood survivors, comprising trauma indexes and emotional reactivity (Strelau and Zawadzki, 2005; Strelau, Kaczmarek, and Zawadzki, 2006). According to the model of relationships between temperament and stress phenomena (see Figure 1), the role of temperament and Figure 2. Interaction Effects Between Exposure to Trauma and Emotional Reactivity in Sample A1 and Sample A2 Notes: TE = trauma exposure; ER = emotional reactivity. 0.9 0.7 0.5 0.3 0.1 –0.1 –0.3 –0.5 –0.7 MAY–JUNE 2012 61 the status of PTSD symptoms may be considered. As expected, emotional reactivity is related to PTSD symptoms but the inter – action with exposure to trauma was also significant. These MVA survivors, who are more emotionally reactive, react to the ac- cident as more traumatic. With the exception of one interaction, all indexes of trauma tend to be more highly related to PTSD symptoms in more emotionally reactive subjects than in less reac- tive people. This refers both to more subjective aspects of trauma as peritraumatic emotions or peritraumatic dissociation and to more objective measures such as the amount of material losses or injuries to the body. While these findings are in accordance with theoretical expectations, the direction of these relationships is unclear. It is also possible that those who experience a more traumatic accident change their temperament characteristics in the aftermath of such an experience. More convincing results would be gathered in prospective studies, although in the traumatic stress domain such research is very rare. In a few studies (e.g., Engelhard, van den Hout, and Kindt, 2003; Gil, 2005) it was stated that temperament traits such as neuroticism or harm avoidance (strongly related to ER; see Strelau and Zawadzki, 1995) mea- sured in the period before trauma occurrence are also significant and positive predictors of PTSD symptoms in the time after a traumatic experience. Thus, in more general terms, emotional reactivity may be considered a personal characteristic, influenc- ing individual “vulnerability” or, in the case of those with low emotional reactivity, “resilience” to trauma (Bonanno, 2004). The main challenge for further analyses is to identify the mechanisms or intermediate variables responsible for the relationship between emotional reactivity and PTSD, besides the hypothetical construct of “arousability” as a psychophysiological basis of temperament traits (Strelau, 1998, 2008). As stated in our previous studies, emotional reactivity can also moderate the process of coping with stress—it is related mostly to coping styles or strategies focused on emotions (Strelau and Zawadzki, 2005; Zawadzki, Kaczmarek, and Strelau, 2009). 3 It is also probable that ER is related to several cognitive characteristics, such as self-efficacy beliefs (£ uszczyñska, Benight, and Cieœ lak, 2009), which are recently under consideration 62 JOURNAL OF RUSSIAN AND EAST EUROPEAN PSYCHOLOGY as main factors influencing human reactions to trauma and facilitating recovery or promoting well-being. It can also be discussed whether PTSD symptoms measured us- ing a whole sample of survivors should be considered as a measure of some kind of symptoms of specific distress in the aftermath of an extremely stressful experience or, in terms of temperament, the stress model, should be treated as a cost of stress and as measures of maladaption in the process of coping. According to the most cur – rent models of development of PTSD and in accordance with the fact that most survivors do not suffer long-term psychopathological consequences, this latter approach should be considered. However, in our study, which analyzed PTSD symptoms as a continuous dimension, this first approach better fits the model. Besides the role of emotional reactivity, exposure to trauma is a good predictor of PTSD symptoms. These findings seem to be unspe- cific to the MVA survivor samples, although it was also established in previous studies on MVA survivors (Blanchard and Hickling, 1999). Surprisingly, threat to life turned out to be the weakest predictor among all of trauma indexes included. This result is rather difficult to explain because in terms of trauma definition based on the PTSD description (APA, 1994), the opposite hypothesis may be expected. The possible interpretation is that threat to life is a very common experience during a traffic accident and because it is so unspecific it is relatively more weakly related to its psychopathological conse- quences. The more detailed features of the experience (e.g., injuries to the body) should be more strongly related. In conclusion, our Polish studies on traumatic stress are consis- tent with international studies and results. However, taking into account temperament traits, which are rarely included in interna- tional studies, may offer more insight into personality and its role in extreme stress situations. Notes 1. In regression analyses, the threat to life was dropped due to the highest correlations of this characteristic with other trauma indexes, especially with injuries (except for the correlation between emotions and dissociation). In the MAY–JUNE 2012 63 two-variable model (ER and particular trauma indexes; see Table 3), however, the influence of threat to life for PTSD as well as its interaction with ER was statistically significant. 2. Three-way and even two-way interactions among trauma indexes were not significant or sample specific, which may suggest the absence of a synergistic effect, and the presence of only an additive effect of trauma characteristics (APA, 1994). 3. Correlation between ER and trauma total index was equal to .17* in sample A1 and .19* in sample A2—for particular indexes: .02 and .06 (threat to life); .12* and .06 (injuries); .09* and .09* (losses); .15* and .20* (dissociation); and .19* and .23* (peritraumatic emotions), respectively, which suggests that emotional reactivity may also influence the perception of trauma severity. These results, however, are only partly consistent with data from flood survivors, for whom the correlations with trauma characteristics (threat, injuries, and losses) were on average equal to .05 and .10 with peritraumatic emotions, and, with dissociation equal to .13 (Zawadzki, Kaczmarek, and Strelau, 2009). References American Psychiatric Association (APA). 1994. DSM IV. Diagnostic and Statistical Manual of Mental Disorders. 3d ed. Washington, DC. Blanchard, E.B., and E.J. Hickling. 1999. “Who Develops PTSD from Motor Vehicle Accidents, and Who Subsequently Recovers?” In The International Handbook of Road Traffic Accidents and Psychological Trauma. Current Understanding, Treatment and Law, ed. Hickling, and Blanchard, pp. 185– 98. Amsterdam: Elsevier Science. Bonanno, G. 2004. “Loss, Trauma, and Human Resilience: Have We Underestimated the Human Capacity to Thrive After Extremely Aversive Events?” American Psychologist, vol. 59, pp. 20-28. Brewin, C.R.; B. Andrews; and J.D. Valentine. 2000. “Meta-Analysis of Risk Factors for Posttraumatic Stress Disorder in Trauma-Exposed Adults.” Journal of Consulting and Clinical Psychology, vol. 68, pp. 748–66. Cohen, J., and P. Cohen. 1983. Applied Multiple Regression/Correlation Analysis for the Behavioral Sciences. Hillsdale, NJ: Erlbaum. Engelhard, I.M.; M.A. van den Hout; and M. Kindt. 2003. “The Relationship Between Neuroticism, Pre-Traumatic Stress, and Post-Traumatic Stress: A Prospective Study.” Personality and Individual Differences, vol. 35, pp. 381–88. Gil, S. 2005. “Pre-Traumatic Personality as a Predictor of Post-Traumatic Stress Disorder Among Undergraduate Students Exposed to Terrorist Attack: A Prospective Study in Israel.” Personality and Individual Differences, vol. 39, pp. 819–27. Kaczmarek, M.; I. KaŸmierczak; and J. Strelau. 2009. “Ekspozycja na traumê oraz cechy osobowoœci jako czynniki modyfikuj¹ce nasilenie symptomów PTSD w grupie pogorzelców: badania rodzinne.” In Konsekwencje psychiczne traumy. Uwarunkowania i terapia, ed. J. Strelau, B. Zawadzki, 64 JOURNAL OF RUSSIAN AND EAST EUROPEAN PSYCHOLOGY and M. Kaczmarek, pp. 110–31). Warsaw: Wyd. Naukowe Scholar. Lis-Turlejska, M. 1998. Traumatyczny stres. Koncepcje i badania. Warsaw: Wydawnictwo Instytutu Psychologii PAN. £uszczyñska, A.; C.C. Benight; and R. Cieœlak. 2009. “Self-Efficacy and Health-Related Outcomes of Collective Trauma. A Systematic Review.” European Psychologist, vol. 14, pp. 49–60. Ozer, E.J.; S.R. Best; T.L. Lipsey; and D.S. Weiss. 2003. “Predictors of Posttraumatic Stress Disorder and Symptoms in Adults: A Meta-Analysis.” Psychological Bulletin, vol. 129, pp. 52–73. Ruscio, A.M.; J. Ruscio; and T.M. Keane. 2002. “The Latent Structure of Posttraumatic Stress Disorder: A Taxonomic Investigation of Reactions to Extreme Stress.” Journal of Abnormal Psychology, vol. 111, pp. 290–301. Strelau, J. 1998. Temperament: A Psychological Perspective. Washington, DC: Plenum Press. ———. 2008. Temperament as a Regulator of Behavior. New York: Eliot Werner. Strelau, J., and B. Zawadzki. 1993. “The Formal Characteristics of Behaviour— Temperament Inventory (FCB-TI): Theoretical Assumptions and Scale Construction.” European Journal of Personality, vol. 7, pp. 313–36. ———. 1995. “The Formal Characteristics of Behaviour—Temperament Inventory (FCB-TI): Validity Studies.” European Journal of Personality, vol. 9, pp. 207–29. ———. 2005. “Trauma and Temperament as Predictors of Posttraumatic Stress Disorder After Disaster.” European Psychologist, vol. 10, pp. 124–35. Strelau, J.; M. Kaczmarek; and B. Zawadzki. 2006. “Temperament as Predictor of Maladaptive Behavior Under Extreme Stress. The Polish Studies of Natural Disasters.” In Progress in Psychological Science Around the World, Vol. 2: Social and Applied Issues, ed. Q. Jing, M.R. Rosenzweig, G. d’Ydewalle, H. Zhang, H.-C. Che, and K. Zhang, pp. 139–58. Hove and New York: Psychology Press. Strelau, J.; B. Zawadzki; W. Oniszczenko; and A. Sobolewski. 2002. “Kwestionariusz PTSD—wersja czynnikowa (PTSD-C). Konstrukcja narzêdzia do diagnozy g³ównych wymiarów zespo³u stresu pourazowego.” Przegl¹d Psychologiczny, vol. 45, pp. 149–76. Zawadzki, B.; M. Kaczmarek; and J. Strelau. 2009. “Reaktywnoœæ emocjonalna a nasilenie objawów zaburzenia stresowego pourazowego u ofiar powodzi: efekt patoplastycznoœci, spektrum, podatnoœci czy komplikacji. In Konsekwencje psychiczne traumy. Uwarunkowania i terapia, ed. J. Strelau, B. Zawadzki, and M. Kaczmarek, pp. 77–104. Warsaw: Wyd. Naukowe Scholar. To order reprints, call 1-800-352-2210; outside the United States, call 717-632-35 35. Copyright of Journal of Russian & East European Psychology is the property of M.E. Sharpe Inc. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder’s express written permission. However, users may print, download, or email articles for individual use.
LRSC
The impact of different diagnostic criteria on PTSD prevalence A comparison of PTSD prevalence using the DSM-IV and ICD-10 PTSD-criteria on a population of 242 Danish social work students MAJA O’CONNOR Department of Psychology MATHIAS LASGAARD University of Aarhus HELLE SPINDLER ASK ELKLIT The diagnostic criteria for PTSD have undergone several changes in the last two decades. This may in part explain the great variance in PTSD prevalence found in existing research. The objective of this study is to investigate the influence of different diagnostic criteria and different combinations of criteria on PTSD prevalence. A sample of 242 Danish social work students (M =29.2 years) completed a list of potentially traumatizing events, major life events and the Harvard Trauma Questionnaire. A considerable difference in PTSD prevalence as a result of different diagnostic criteria of PTSD was found. Future meta-analyses and reviews of PTSD prevalence must take into account the impact of changing criteria on prevalence. Clinicians also need to address this issue when assessing PTSD. Key words: Traumatic events, diagnostic criteria of PTSD, PTSD prevalence, DSM-IV, ICD-10. Correspondence: Maja O’Connor, Department of Psychology, University of Aarhus, Jens Chr. Skous Vej 4, 8000 Aarhus C, Denmark. Tel.: +45 8942 4926, Fax: +45 8942 4901. E-mail: [email protected] The experience of a traumatic event involving actual or threatened death, or serious injury may lead to the development of PTSD. In turn, receiving a PTSD diagnosis may provide a client with the option of psychological or medical care, and compensation. However, the same person may be diagnosed differently depending on which general diagnostic criteria of PTSD is used and which specific criteria, from for example DSM-IV, has been selected. Clinicians and researchers studying psychological trauma need to be aware of the potential vari- ance in diagnostic outcome as a product of the diagnostic criteria used. Prevalence of PTSD Many studies have aimed to establish the prevalence and risk of PTSD for specific traumatic events, e.g. violence (e.g. Boney-Mccoy & Finkelhor, 1995; ARTICLE Nordic Psychology, 2007, 59 (4) 317-331 This document is copyrighted by the American Psychological Association o r one of its allied publishers. This article is intended solely for the personal use of the individual u ser and is not to be disseminated broadly. 318 Maja O’Connor et al. NP, 2007 (3) Seedat, Nyamai, Njenga, Vythilingum & Stein, 2004), childhood abuse (Libby, Orton, Novis, Beals & Manson, 2005) and war (Khamis, 2005). In contrast, studies investigating a broad range of traumatic events and the associated risk of PTSD are not very common. Four national probability samples (Kessler, Sonnega, Bromet, Hughes & Nelson, 1995; Kessler, Berglund, Demler, Jin, Marikangas, et al., 2005; Perkonigg, Kessler, Storz & Wittchen, 2000; Frans, Rimö, Åberg & Fredrikson, 2005) and two epidemiological studies (Bernat, Ronfeldt, Calhoun & Arias. 1998; Hepp, Gamma, Milos, Ajdacic-Gross, et al., 2006) have explored a broad range of traumatic events in adult populations (based on a September 2006 PsycInfo search using different combinations of search keywords such as; PTSD, prevalence, ICD 10, DSM IV, epidemiologic study, probability study). Similarly, one national probability study (Elklit, 2002) and one epidemiological study (Costello, Erkanli, Fairbank & Angold, 2002) have explored the prevalence of traumatic events in adolescent populations (based on same search as above). These studies show that the majority of the adults and adolescents sampled had been exposed to one or more traumatic events, resulting in a considerable number of PTSD cases. The rates of prevalence found in the above studies show great variation regard- less of age. The number of participants reporting one or more events ranges from 21 % to 87 % and there is large variation in the prevalence reported for specific events and most common event experienced. Moreover, the prevalence of PTSD shows great variation across studies. The lifetime risk of PTSD ranges from 0 % to 9 %, and varies markedly across different types of trauma, although most studies find sexual abuse/rape to be associated with an increased risk of PTSD. In line with the recent findings of Foa & Tolin (2006) most studies show that females are more likely to suffer from PTSD, despite males reporting more exposure to traumatic events than females. The reported differences in event-rates and PTSD prevalence may be related to methodological differences, e.g. the degree of specificity of the measured events, the number of events investigated, differences in data collection methods, demographics, and cultural and community related variables. Furthermore, dif- ferent problems such as the effect of the inclusion of the emotional impact, the A2 criterion (a subjective component of PTSD involving a response of intense fear, helplessness, or horror) or the diagnostic boundaries of PTSD in DSM IV are associated with the use of the PTSD diagnosis and have been widely discussed (e.g. Norman, Stein & Davidson, 2007; Brewin, 2005; Creamer, McFarlane & Burgess, 2005; Schützwohl & Maercker, 1999; O’Donohue & Elliot, 1992). This document is copyrighted by the American Psychological Association o r one of its allied publishers. This article is intended solely for the personal use of the individual u ser and is not to be disseminated broadly. 319 The impact of different diagnostic criteria on PTSD prevalence NP, 2007 (3) Development of the diagnostic criteria of PTSD The diagnostic criteria of PTSD have developed over time. Prior to the intro- duction of the PTSD diagnosis in DSM-III (APA, 1980), trauma-related symp- tomatology was recognized in DSM-I (1952) as Gross Stress Reaction, and in DSM-II as Adjustment Reaction of Adult Life (Wilson, 1994). In DSM-III different syndromes caused by various traumatic events were finally combined into one diagnosis, Post Traumatic Stress Disorder. To qualify for the diagnosis of PTSD the existence of a stressor “that would evoke significant symptoms of distress in almost anyone had to be identifiable” (criteria 1). The remaining criteria were grouped into three symptom sections consisting of: 2) re-experiencing the traumatic event (recurrent and intrusive recollections of the event, recurrent dreams, and/or feeling as if the traumatic event is re-occurring), 3) numbing or detachment (diminishes interest in significant activities, feeling detachment or estrangement from others, and/or constricted affect) and 4) symptoms not pres- ent before the traumatic event (hyper alertness or exaggerated startle response, sleep disturbance, survivor guilt, memory impairment or trouble concentrating, avoidance of activities that arouse recollection of the traumatic event, and/or intensification of symptoms when reminded of the traumatic event). At least one symptom from section 2 and 3 and at least two symptoms from section 4 must be present to qualify for the PTSD diagnosis. The diagnostic criteria were thoroughly revised in DSM-III-R (APA, 1987) adding a new component (E) of duration of symptom clusters B, C, and D, for at least one month and with onset in the immediate aftermath of the traumatic event (B, C, and D were a revision of section 2, 3, and 4 in DSM-III). Furthermore now, the stressor criterion (A) stated that the person must have experienced “an event outside the range of usual human experience that would be markedly distressing to almost anyone”. The symptom clusters in DSM-III-R consisted of: (B) persistent re-experiencing the traumatic event (recurrent and intrusive distress- ing recollections of the event, recurrent distressing dreams of the event, sudden acting or feeling as if the traumatic event is reoccurring and/or intense psycho- logical distress when exposed to situations reminding of the traumatic event): (C) persistent avoidance or numbing in relation to the traumatic event (efforts to avoid thoughts or feelings associated with the trauma, efforts to avoid activities or situations evoking recollection of the trauma, inability to recall an important aspect of the trauma, markedly diminished interest in significant activities, feeling detachment or estrangement from others, restricted affect and or a sense of fore- shortened future), (D) persistent arousal not present before the trauma (difficulties falling or staying asleep, irritability or outbursts of anger, problems concentrating, hypervigilance, exaggerated startle response and/or physiologic reactivity when This document is copyrighted by the American Psychological Association o r one of its allied publishers. This article is intended solely for the personal use of the individual u ser and is not to be disseminated broadly. 320 Maja O’Connor et al. NP, 2007 (3) exposed to trauma related events). At least one symptom from cluster B, at least three symptoms from cluster C, and at least two symptoms from cluster D must be present to qualify for diagnosis (APA, 1987). DSM-IV (APA, 1994) enlarged the definition of stressors into two subcriteria (A1 and A2) and a functional criterion of clinically significant distress or impairment (F). Relatively little research on criterion (F) has yet been published. According to DSM-IV (APA, 1994) the diagnosis of PTSD requires the exposure to a traumatic event that involves: (A1) Actual or threat of death or serious injury, or a threat to the physical integrity of self or others, and (A2) involving a response of intense fear, helplessness, or horror. Moreover, the diagnosis requires (B) persistent re- experiencing of the traumatic event, (C) persistent avoidance of stimuli associated with the trauma and numbing of general responsiveness, and (D) persistent symp- toms of increased arousal. Finally, (E) the full symptoms must be present for more than 1 month, and (F) the disturbance must cause clinically significant distress or impairment in social, occupational, or other important areas of functioning. At least one symptom from cluster B, at least three symptoms from cluster C, and at least two symptoms from cluster D must be present to qualify for diagnosis. In the International Classification of Diseases 9 (ICD-9) from 1978, the World Health Organization (WHO) recognized the possible emotional problems follow- ing traumatic experiences by including two diagnoses (Acute Reaction to Stress and Adjustment Reaction) (Joseph, Williams & Yule, 1997). The WHO (1993) included a diagnosis of PTSD in their most recent edition, ICD-10, as “a reaction to severe stress and adjustment disorders”. To diagnose the disorder, identification of (A) a stressor that is “likely to cause severe distress in almost anyone” is necessary (e.g. disaster, combat, rape, terrorism, violent death of others). In addition the following symptoms are required: (B) repetitive symp- toms of re-experiencing the traumatic experience (intrusive recollection or re- enactment in memories, dreams, or imagery) or severe discomfort when reminded of the traumatic experience, (C) actual or preferred avoidance of reminders of the traumatic experience, (D) either (D1) partially or complete inability to recall important aspects of the traumatic event or (D2) two of the following symptoms of increased arousal (difficulty in falling or staying asleep, irritability or outbursts of anger, difficulty concentrating, hypervigilance, exaggerated startle response), and (E) the criteria B, C, and D must be met within 6 months of the traumatic event. Several studies and reviews have discussed differences and similarities between the general diagnostic criteria of PTSD in DSM-IV and ICD-10. Empirical stud- ies almost exclusively apply DSM-IV criteria in their research. In contrast, clinical diagnosis and practice in many countries is based on ICD-10. However, when applying evidence-based treatment of PTSD, this evidence almost exclu- sively builds on research using DSM-IV criteria for the diagnosis in question. This document is copyrighted by the American Psychological Association o r one of its allied publishers. This article is intended solely for the personal use of the individual u ser and is not to be disseminated broadly. 321 The impact of different diagnostic criteria on PTSD prevalence NP, 2007 (3) Consequently, there is a risk that using treatment strategies based on DSM-IV results in lack of coherence between evidence-based clinical practice and ICD-10 diagnosis. General consensus holds that there are great similarities, but also important differences between the two systems, mainly that ICD-10 does not emphasize avoidance and increased arousal to the same extent as DSM-IV, and that DSM-IV includes criteria of more than 1 months duration (E) and impairment of function (F) that ICD-10 does not (Joseph et al., 1997; Peters, Slade, and Andrew, 1999; Andrews, Slade, and Peters, 1999; Lundin & Lofti, 1996; Lopez-Ibor, 2002; Peters, Issakidis, Slade & Andrews, 2005). It is noteworthy that only two studies investi- gating differences in identification of PTSD cases using DSM-IV and ICD-10 have been identified. These studies found that only 37% of cases identified through ICD-10 (PTSD prevalence = 7%) also fulfilled the criteria of DSM-IV (PTSD preva- lence = 3 %), while on the other hand 85% of cases fulfilling DSM-IV criteria also fulfilled ICD-10 criteria (Andrews et al., 1999; Peters et al., 1999). Peters and colleagues suggested that this difference appears because DSM-IV requires fulfillment of a functional criterion (F), and put more emphasis on symptoms of avoidance or numbing (C) compared to ICD-10 (Peters et al., 1999). Moreover, a study focusing on gender differences found twice as many cases of PTSD using ICD-10 compared to DSM-IV (Peters et al., 2005). Diagnostic criteria and prevalence of PTSD Epidemiological studies have used different diagnostic criteria that likely influ- ence the reported prevalence of PTSD and hence possibly explain some of the differences (see Table 1). The results of a cohort study (n=367) by Hepp and colleagues (2006) in which they found no cases at all of full PTSD is worth mentioning. The participants in this study reported whether or not they, dur- ing the last 12 months, had experienced or witnessed an event that involved actual or threatened death, serious injury or threat to the physical integrity of others – corresponding closely to the DSM-IV criteria A1. Participants were only asked to categorize the type of event if they answered affirmative to this question. The estimated prevalence of lifetime exposure to potentially traumatic events was 28 %, which may be explained by the very broad formulation of this question in terms not easily understandable to the population investigated. Another explanation could be that specific categories of events, as for example the categorizations developed by Kessler et al. (1995) are more likely to trigger recognition of specific, potentially traumatic events than the broader approach used by Hepp and colleagues (2006). This does not, however, explain why none of the 128 individuals fulfilling criteria A1 met all the remaining criteria This document is copyrighted by the American Psychological Association o r one of its allied publishers. This article is intended solely for the personal use of the individual u ser and is not to be disseminated broadly. 322 Maja O’Connor et al. NP, 2007 (3) of PTSD (Hepp et al., 2006). One reason may be that the participants, con- sistent with, for example, Perkonigg and colleagues (2000), had to meet all 6 specific diagnostic criteria (A-F) to qualify for the diagnosis of PTSD. However, in a Swedish probability study (n=1824) Frans and colleagues (2005) found a lifetime prevalence of 5.6 % using all 6 specific criteria of PTSD and using yes/ no answers to identify symptoms, hence possibly increasing the effect. In comparison, the remaining studies described in Table 1 found a lifetime prevalence of PTSD ranging from 4-9 %. This may be explained by the use of less specific DSM-IV criteria for PTSD or different general diagnostic criteria (DSM-III-R). Table 1 Lifetime prevalence of PTSD and general and specific criteria in national populations Lifetime prevalence (All partici- pants)Prevalence in exposed participants (one or more potentially traumatic events reported)General and specific diagnostic criteria applied Kessler et al., 19957.8 % 11.8 % DSM-III-Ra Kessler et al., 20056.8 % – DSM-IVa Elklit, 20029 % 10.3% DSM-IV A1, B, C, & D Bernat et al. 19984 % 12 % DSM-IV A1, A2, B, C, & D Hepp et al. 20060 % 0 % DSM-IV A1, A2, B, C, D, E, & F Perkonigg et al., 20001.3 % 7.8 % DSM-IV A1, A2, B, C, D, E, & F Frans et al., 20055.6 % 6.9 % DSM-IV A1, A2, B, C, D, E, & F a No information available about specific criteria used. While trauma and PTSD in non-clinical adult populations have been investigated (e.g. Kessler et al., 1995; Kessler et al 2005; Perkonigg et al., 2000; Frans et al., 2005), few studies have investigated the effect of different diagnostic criteria applied to the same population. Three studies have found significant variations in PTSD prevalence according to the different diagnostic criteria (Bernat et al. 1998; Peters et al., 2005; Peters et al., 1999). Concurrently, the prevalence of PTSD varies widely between studies. Different diagnostic criteria and configurations of symptoms may partly explain variations found in studies of PTSD prevalence, This document is copyrighted by the American Psychological Association o r one of its allied publishers. This article is intended solely for the personal use of the individual u ser and is not to be disseminated broadly. 323 The impact of different diagnostic criteria on PTSD prevalence NP, 2007 (3) both in non-clinical and clinical populations, emphasizing the relevance of studying this issue further. Aim of the study The aim of the present study was to explore the impact of different diagnostic criteria on PTSD prevalence in an adult non-clinical population. The hypotheses investigated are: 1) The diagnostic tool used influences the PTSD prevalence, 2) Introducing additional specific DSM-IV criteria for PTSD reduces the prevalence rate of PTSD. Material and methods Participants The sample for the current study consisted of 242 adult students with a mean age of 33 years (SD = 11.26; range 16-61 years). There were 209 (86 %) females and 33 (14 %) males. Forty two percent of the students lived alone (unmarried, divorced or widowed), while 56 % lived together with a partner (married or cohabiting). 50 % of the students had children. The average length of education was 13 years (SD 3.1; range 7-24 years). Procedures The first author introduced the study to the students and was available for clarifying questions while the participants completed the questionnaire in the classroom. Participation was voluntary and the response rate was 99%. The par- ticipants were recruited from three different schools of intermediate educational level located in Aarhus, Denmark. The participants were selected because of their status as adult social worker students, of which the majority had a non- university background. A Tukey B post hoc analysis, that allow statistically reli- able identification of differences between two groups or more (Pallant, 2005), showed few differences between the three groups on parameters of gender, age and range of traumatic events and life events. Measures The first part of the questionnaire contained socio-demographic questions about gender, age, education and family conditions. Following that, two types of stressors, traumatic experiences ad modum Kessler and colleagues (1995) and distressing life events were investigated. Distressing life-events were reported by answering the following question: “Within the last year have you experienced This document is copyrighted by the American Psychological Association o r one of its allied publishers. This article is intended solely for the personal use of the individual u ser and is not to be disseminated broadly. 324 Maja O’Connor et al. NP, 2007 (3) major life-events/changes?”. Subsequently, the students were asked about the experience of different traumatic events (see Table 2). Table 2 Trauma and Life Events According to Exposure and Prevalence of PTSD (fulfilling criteria A1, B, C, and D of DSM-IV) EventFrequency (%)Choice of most distressing event (%)Relative risk (%) PTSDSub clinical PTSD Traumatic events Accident 20 38 38 32 Shock by some one close being exposed to a traumatic event 16 21 51 35 Threat of violence 15 18 38 34 Serious illness 13 28 50 23 Childhood abuse 14 52 75 14 Violent assault 10 24 41 28 Witness other people getting injured or killed 9 3 38 29 Rape 2 86 100 0 Recent life events Move of residence 17 13 54 20 Divorce/break up with a partner 6206414 Change of employment or education 5 6 46 8 Getting fireda 238– a = less than 5 cases Harvard Trauma Questionnaire-Part IV (Mollica, Capsi-Yavin, Bollini, & Truong, 1992) was used to estimate the occurrence of PTSD. HTQ consists of 30 items, rated on a 4-point Likert scale (1 = not at all; 4 = very often). Sixteen items relate to the three core clusters in PTSD in DSM-III-R: intrusion, avoidance, and arousal. In the present study, participants rated the HTQ on the basis of the most stressful life or traumatic event experienced. Participants answered HTQ in relation to their reaction in the time immediately after the stressful event. Only scale items ≥ 3 on HTQ were considered for a PTSD diagnosis. Recognition of a sub clinical level of PTSD was given if the respondent met two of the three criteria. The Danish version of the HTQ has been found to be a reliable and valid mea- sure (Bach, 2003). Moreover, HTQ ratings according to the DSM-III-R diagnostic This document is copyrighted by the American Psychological Association o r one of its allied publishers. This article is intended solely for the personal use of the individual u ser and is not to be disseminated broadly. 325 The impact of different diagnostic criteria on PTSD prevalence NP, 2007 (3) criteria of PTSD showed an 88% concordance with interview based estimates of PTSD (Mollica et al., 1992). The internal consistency of the PTSD scale and subscales in the present study was good for the HTQ (Total score: α = .95; intru- sion: α = .81; avoidance: α = .73; arousal: α = .73). Furthermore, the following four single questions relating to the chosen event were included in the study with the purpose of establishing whether the stressful event selected when filling in the HTQ met the A1 and A2 criteria included in DSM-IV-TR. 1) Were you in mortal danger while it happened? 2) Were others in mortal danger? 3) Were you injured? 4) Did you feel helpless or horrified? Moreover, the participants were asked to report how much each symptom dis- turbed her/him during the last month to investigate present PTSD symptoms. Statistical Analysis The sample used in this study was extracted from a larger study. Prior to data analysis, we excluded a fourth group, consisting of 86 predominantly young, male, trainee-craftsmen, due to inadequate data quality. To fulfill the criteria of DSM-III, the following demands had to be met: (1) The participant reported one or more traumatic events; the participant reported ≥ 3 on: (2) one or more intrusion items (HTQ 1-3), (3) on one or more numbing items (HTQ 4, 5, 13) and (4) on two or more items of symptoms not present before the traumatic event (HTQ 6-8, 11, 12, 16, 20). To fulfill the five criteria of DSM-III-R, the following was computed: (A) The participant reported one or more traumatic events; the participant reported ≥ 3 on: (B) one or more intrusion items (HTQ 1-3), (C) three or more symptoms of persistent avoidance or numbing (HTQ 4, 5, 11-15), (D) two or more symptoms of persistent arousal not present before the trauma (HTQ 6-10, 16). To fulfill the DSM-IV criteria of PTSD the participant reported (A1) one or more traumatic events and (A2) a sense of helplessness or horror in relation to the traumatic event. The participant reported one or more traumatic events; the participant reported ≥ 3 on (B) one or more intrusion items (HTQ 1-3, 16) item 16 containing both the physiological and psychological stress of being reminded of the event; (C) three or more symptoms of persistent avoidance or numbing (HTQ 4, 5, 11-15) and (D) two or more symptoms of persistent arousal not pres- ent before the trauma (HTQ 6-10). HTQ18 “Difficulties with carrying out work or daily functions” was defined as the DSM-IV criterion (F) where the disturbance must cause clinically significant distress or impairment in social, occupational, or other important areas of functioning. Criterion (F) was defined as fulfilled if the participant reported a score ≥ 3 on this item. This document is copyrighted by the American Psychological Association o r one of its allied publishers. This article is intended solely for the personal use of the individual u ser and is not to be disseminated broadly. 326 Maja O’Connor et al. NP, 2007 (3) To fulfill the four first criteria of ICD-10 PTSD, the following algorithm was applied: (A) The participant reported one or more traumatic events; the par- ticipant reported ≥ 3 on (B) one or more symptoms of persisting intrusion or re-experiencing the traumatic event (HTQ 1-3 or 16); (C) actual or preferred avoidance of circumstances associated with the stressor (HTQ 11 or 15); (D) inability to recall important aspects of the trauma (HTQ 12) or two out of five symptoms of increased arousal (HTQ 6-10). The last criteria (E) where B, C, and D must be met within 6 months of the traumatic event could not be included in the computation. Frequency analysis or Chi-square analyses were computed of the concordance between participants identified as fulfilling the diagnostic criteria of PTSD in ICD-10 compared to DSM-IV and DSM-IV compared to ICD-10. Only par- ticipants with valid scores on both parameters were included. All analyses were carried out using SPSS, version 13. Results A total of 87 % of the students reported at least one life event or at least one traumatic experience (see Table 2). Eighty percent of the students reported one or more traumatic event (one event = 31 %, two events = 23 %, three or more events = 26 %). The average number of experienced traumatic events was 1.7 (SD = 1.5), and the most common events reported were bereavement, shock by someone close being exposed to a traumatic event, threat of violence, and accident. Twenty-nine percent of the students reported one or more distressing life events within the last year (one event = 23 %, two events = 6 %, three or more events = 1 %). The average number of distressing life events per student was 0.4 (SD = 0.6), and the most commonly reported life event was change of residence, and divorce/separation. The PTSD prevalence was remarkably high in those reporting certain major life-events within the last year. Divorce or break-up with a partner was the highest ranking (64% with PTSD), followed by change of residence (54%), and change of employment or education (46%). The average number of experienced events was higher among participants with PTSD than exposed participants without PTSD (2.6 vs. 1.9 events). Table 3 illustrates that the diagnostic tool used influenced the PTSD prevalence rate. The largest difference was found between ICD-10 with a PTSD prevalence of 35% and DSM-IV including criteria A1, A2, B, C, D, E, and F with a PTSD prevalence of 11%. Furthermore, as expected, introducing additional DSM-IV criteria reduced the PTSD prevalence. This document is copyrighted by the American Psychological Association o r one of its allied publishers. This article is intended solely for the personal use of the individual u ser and is not to be disseminated broadly. 327 The impact of different diagnostic criteria on PTSD prevalence NP, 2007 (3) Table 3 Prevalence of lifetime PTSD according to different diagnostic criteria and different con- figurations of specific criteria within DSM-IV PTSD-prevalenceSub-clinical PTSD (2 out of 3 criteria: B, C, and D) ICD-10 A, B, C, & D35 % (n=84) 17 % (n=42) DSM-III-R A, B, C, & D30 % (n=73) 16 % (n=39) DSM-III 1, 2, 3 & 429 % (n=70) 14 % (n= 33) DSM-IV B, C, D, & E25 % (n=61) 17 % (n=42) DSM-IV A1, B, C, D, & E22 % (n=54) 16 % (n=39) DSM-IV A2, B, C, D, & E20 % (n=48) 15 % (n=35) DSM-IV A1, A2, B, C, D, & E17 % (n=42) 14 % (n=33) DSM-IV A1, A2, B, C, D, E, & F11 % (n=26) 4 % (n=10) Using all 6 DSM-IV criteria of PTSD, 26 participants qualified for the full PTSD diagnosis. Ninety two percent of these (24 participants) also qualified for the ICD-10 diagnosis of PTSD using criteria A, B, C, and D. Eighty one participants qualified for the ICD-10 diagnosis of PTSD. Three participants were excluded because of missing scores in the DSM criteria. Out of the 81 participants, only 30 percent (24 participants) qualified for the full diagnosis of PTSD according to DSM-IV (criteria A1, A2 B, C, D, E, and F). Discussion In line with findings from other studies (e.g. Bernat et al., 1998) a high num- ber (80%) of participants in the present study reported at least one potentially traumatic event. PTSD prevalence was found to vary considerably according to which diagnostic tool was used. The PTSD prevalence according to ICD-10 was 35% while according to DSM-IV using similar specific criteria (A1, B, C, D, and E) the PTSD prevalence was 22% (see Table 3). The results indicate that within the same population the diagnostic tool chosen may have a strong impact on the number of cases of PTSD identified. This may explain why the concor- dance between PTSD in DSM-IV versus ICD-10 was only 30%. In addition, This document is copyrighted by the American Psychological Association o r one of its allied publishers. This article is intended solely for the personal use of the individual u ser and is not to be disseminated broadly. 328 Maja O’Connor et al. NP, 2007 (3) a lack of congruency between practice and research poses a serious problem in countries that use the ICD-10 in clinical work as the empirical evidence is based DSM PTSD. Such in-congruency may result in PTSD losing its credibility as a meaningful and applicable diagnosis in clinical work in countries using ICD-10. In line with the findings by Peters et al. (2005) the present study found a much lower concordance between PTSD in ICD-10 compared to DSM-IV (30 % concordance) and PTSD in DSM-IV compared to ICD-10 (91% concordance). The DSM-IV PTSD prevalence was reduced as additional specific diagnostic criteria were added. This result may seem common sense, but the fact is that several studies only specify the general diagnostic criteria used (e.g. ICD-10 or DSM-IV) without specifying exactly how they operationalized these criteria. As mentioned, large variation has been found in PTSD prevalence as defined by DSM-IV criteria both in clinical and non-clinical samples. One of the reasons for this variability may be the fact that different specific criteria have been applied, but have not been clearly operationalized when reporting the results. This poses an obstacle when reviewing PTSD prevalence. According to the results of this study, it is necessary to take both the general and specific DSM criteria applied into consideration before comparing results from different studies. The main problem is that the applied specific DSM-IV criteria, apart from A1 and A2, are rarely operationalized in studies of PTSD prevalence. Another is that the significant changes in all general DSM criteria variables from 1980 to 1994 must be taken into consideration when comparing results from studies using different general criteria. Obviously, this makes reviewing the literature difficult. To avoid the risk of drawing conclusions based upon incomplete information, future review work on estimation of PTSD prevalence needs to take the variance in PTSD prevalence produced by unspecified and varying diagnostic criteria into account. Also existing meta-analyses and reviews should be assessed with this in mind (e.g. Tolin & Foa, 2006). In conclusion, when investigating PTSD prevalence it is crucial that researchers pay careful attention to the applied diagnostic criteria when performing reviews and meta-analysis. Furthermore researchers must be urged to specify the precise PTSD-criteria used when reporting studies involving PTSD prevalence. The high PTSD prevalence in the present study (see Table 2) might be explained by the fact that many of the participants recently left home or got divorced, moved into their first own residence and started their further education in a new field. This type of event may cause stress, but should not lead to PTSD, as the DSM-IV criteria A1 is not fulfilled. Even so, in a Dutch population including 832 adults, Mol, Arntz, Metsemakers, Dinant, Vilters-Van, et al. (2005) found that PTSD scores were higher after life events than after traumatic events from the past 30 years. In line with this, other studies have reported significant associations This document is copyrighted by the American Psychological Association o r one of its allied publishers. This article is intended solely for the personal use of the individual u ser and is not to be disseminated broadly. 329 The impact of different diagnostic criteria on PTSD prevalence NP, 2007 (3) between PTSD symptoms and 1) the exposure of farmers to an epidemic of foot and mouth disease causing a mass cull of livestock (Olff, Koeter, Van Haaften, Kersten, & Gersons, 2005), 2) extreme pressure including public humiliation in the work setting (Ravin & Boal, 1989), 3) victims of bullying at work (Mikkelsen & Einarsen, 2002), and 4) parental divorce/an absent parent (Elklit, 2002; Joseph, Mynard & Mayall, 2000). In sum, several studies have reported PTSD symptoma- tology in cases lacking a DSM-IV adequate A1 stressor. One possible reason for these findings may be that some stressors, not including actual or threat of death or serious injury, or a threat to the physical integrity of self or others, pose a threat to identity (Brewin, 2003; Mol et al. 2005). A threat to the psychological or social integrity of a person is in some cases comparable to the threat of the physical integrity, included in the present diagnostic criteria. Another potential explanatory factor could be that accumulation of traumatic events experienced was related to PTSD prevalence, since the average number of experienced events was higher among participants with PTSD than among exposed participants without PTSD. In a nationally representative sample of Danish adolescents (Shevlin & Elklit, in press) latent class analysis was used to identify clusters or latent classes of events. In addition, the relationships between the latent classes and living arrangements and diagnosis of post-traumatic stress disorder (PTSD) were estimated. A three-class solution was found to be the best description of multiple adverse life events, and the three classes were labelled ‘Low Risk’, ‘Intermediate Risk’, and ‘High Risk’. The High Risk group were found to have a relatively high likelihood of experiencing multiple traumas and were 13 times more likely to have a PTSD diagnosis. The results of the current study should be interpreted with caution. Firstly, the study is cross-sectional and retrospective. Secondly, the high PTSD prevalence found could be an indication of the fact that our sample was non-representative. On the other hand, the sample of social work students represents a group that may be more similar to the general population than undergraduate university students. Our results cannot be generalized to other populations. Despite the limitations, the study explores an understudied issue in the field of traumatic stress. To summarize, the results from this study indicate that extreme care should be taken regarding both the general and specific diagnostic criteria used when assessing PTSD prevalence. Often only the general criteria (ICD-10 or DSM-IV) and the A1 and A2 criteria of DSM-IV are operationalized, making meta-analysis and reviews on PTSD prevalence difficult to perform in a satisfactory manner. When reporting future research on PTSD prevalence a precise description of which specific DSM-IV criteria are used and operationalized is required. In time, this could lead to consensus on the use and operationalization of PTSD criteria in empirical work and clinical practice. In addition, future studies should also con- This document is copyrighted by the American Psychological Association o r one of its allied publishers. This article is intended solely for the personal use of the individual u ser and is not to be disseminated broadly. 330 Maja O’Connor et al. NP, 2007 (3) sider the possibility of alternative structures of PTSD symptomatology including a widening of the criteria to encompass situations that might threaten the current life conditions and psychological integrity of the person (see Brewin, 2003). REFERENCES APA (1980). Diagnostic and statistical manual of mental disorders. Third Edition. Washington, DC: American Psychiatric Association. APA (1987). Diagnostic and statistical manual of mental disorders. Third edition, revised. Washington, DC: American Psychiatric Association APA (1994). Diagnostic and statistical manual of mental disorders. Fourth edition. Washington, DC: American Psychiatric Association Andrews, G., Slade, T., & Peters, L. (1999). Classification in psychiatry: ICD-10 versus DSM-IV. British Journal of Psychiatry, 174, 3-5. Bach, M.E., (2003). En empirisk belysning og analyse af „Emotional Numbing“ som en eventuel selvstændig faktor i PTSD. Psykologisk Studieskriftserie 6(1), 1-200. Bernat, J., Ronfeldt, H. M., Calhoun, K. S., & Arias, I. (1998). Prevalence of traumatic events and peritraumatic predictors of posttraumatic stress symptoms in a nonclinical sample of college students. Journal of Traumatic Stress, 11(4), 645-664. Brewin, C. (2005). Systematic review of screening instruments for adults at risk of PTSD. Journal of Traumatic Stress, 18(1), 53-62. Costello, E. J., Erkanli, A., Fairbank, J. A., & Angold, A. (2002). The prevalence of potentially trau- matic events in childhood and adolescence. Journal of Traumatic Stress, 15(2), 99-112. Cramer, M., McFarlane, A. & Burgess, P. (2005). Psychopatology following a trauma: The role of subjective experience. Journal of Affective Disorders, 86, 175-182. Elklit, A. (2002). Victimization and PTSD in a Danish national youth probability sample. Journal of the American Academy of Child & Adolescent Psychiatry, 41(2), 174-181. Frans, Ö., Rimmö, P.-A., Åberg, L., & Fredrikson, M. (2005). Trauma exposure and post-traumatic stress disorder in the general population. Acta Psychiatrica Scandinavica, 111(4), 291-299. Hepp, U., Gamma, A., Milos, G., Eich, D., Ajdacic-Gross, V., Rössler, W., Angst, J., Schnyder, U. (2006). Prevalence of exposure to potentially traumatic events and PTSD: The Zurich Cohort Study. European Archives of Psychiatry and Clinical Neuroscience, 256(3), 151-158. Joseph, S., Williams, R., & Yule, W. (1997). Understanding post-traumatic stress. A psychosocial perspective on PTSD and treatment. UK: Wiley. Joseph, S., Mynard, H. & Mayall, M. (2000). Life-Events and Post-traumatic Stress in a Sample of English Adolescents. Journal of Community & Applied Social Psychology, 10, 475-482. Kessler, R. C., Berglund, P., Demler, O., Jin, R., Merikangas, K. R., & Walters, E. E. (2005). Lifetime prevalence and age-of-onset distributions of DSM-IV disorders in the national comorbidity survey replication. Archives of General Psychiatry, 62(6), 593-602. Kessler, R. C., Chiu, W.T., Demler, O., & Walters, E. (2005). Prevalence, severity, and comorbid- ity of 12-month DSM-IV disorders in the national comorbidity survey replication. Archives of General Psychiatry, 62(6), 617-627. Kessler, R. C., Sonnega, A., Bromet, E., Hughes, M., & Nelson, C. B. (1995). Posttraumatic stress disorder in the National Comorbidity Survey. Archives of General Psychiatry, 52(12), 1048-1060. Khamis, V. (2005). Post-traumatic stress disorder among school age Palestinian children. Child Abuse & Neglect, 29, 81-95. Libby, A., Orton, H., Novis, D., Beals, J. & Manson, S. (2005). Childhood physical and sexual abuse and subsequent depressive and anxiety disorders for two American Indian tribes. Psychological Medicine 35(3), 329-340. López-Ibor, J. J. (2002). The classification of stress-related disorders in ICD-10 and DSM-IV. Psychopathology, 35, 107-111. This document is copyrighted by the American Psychological Association o r one of its allied publishers. This article is intended solely for the personal use of the individual u ser and is not to be disseminated broadly. 331 The impact of different diagnostic criteria on PTSD prevalence NP, 2007 (3) Lundin, T., & Lotfi, M. (1996). Posttraumatic stress disorder in DSM-III-R, DSM-IV, and ICD-10: A comparison and evaluation of the significance of the respective diagnostic criteria. Northern Journal of Psychiatry, 50, 11-15. Mikkelsen, E. & Einarson, S. (2002), Basic assumptions and symptoms of post-traumatic stress among victims of bullying at work, European Journal of Work and Organizational Psychology, 11(1), 87-111. Mol, S. S. L., Arntz, A., Metsemakers, J. F. M., Dinant, G.-J., Vilters-Van Montfort, P. A. P., & Knottnerus, J. A. (2005). Symptoms of post-traumatic stress disorder after non-traumatic events: Evidence from an open population study. British Journal of Psychiatry, 186(6), 494-499. Norman, S., Stein, M. & Davidson, J. (2007). Profiling Posttraumatic Functional Impairment. The Journal of nervous and Mental Disease, 195(1), 48-53. Mollica, R., Caspi-Yavin, Y., Bollini, P., Truong, T., Tor, S. & Lavelle, J. (1992). The Harvard Trauma Questionnaire. Validating a Cross-Cultural Instrument for Measuring Torture, Trauma an Posttraumatic Stress Disorder in Indochinese Refugees. The Journal of Nervous and mental Disease, 180(2), 111-118. O’Donohue, W., & Elliott, A. (1992). The current status of post-traumatic stress disorder as a diag- nostic category: Problems and proposals. Journal of Traumatic Stress, 5(3), 421-439. Olff, M., Koeter, M., Van Haaften, E., Kersten, P. & Gersons, B. (2005). Impact of a foot and mouth disease crisis on post-traumatic stress symptoms in farmers. British Journal of psychiatry, 186, 165-166. Pallant, J. (2005). SPSS Survival Manual. 2nd edition. A step by step guide to analysis using SPSS for Windows (Version 12). UK: Open University Press. Perkonigg, A., Kessler, R. C., Storz, S., & Wittchen, H.-U. (2000). Traumatic events and post- traumatic stress disorder in the community: Prevalence, risk factors and comorbidity. Acta Psychiatrica Scandinavica, 101(1), 46-59. Peters, L., Issakidis, C., Slade, T., & Andrews, G. (2005). Gender differences in the prevalence of DSM-IV and ICD-10 PTSD. Psychological Medicine, 35, 1-9. Peters, L., Slade, T., & Andrews, G. (1999). A comparison of ICD10 and DSM-IV criteria for post- traumatic stress disorder. Journal of Traumatic Stress, 12(2), 335-343. Ravin, J. & Boal, C. (1989). Post-Traumatic Stress Disorder in the Work Setting: Psychic Injury, Medical Diagnosis, Treatment and Litigation. American Journal of Forensic Psychiatry, 10(2), 5-23. Schützwohl, M., & Maercker, A. (1999). Effects of varying diagnostic criteria for posttraumatic stress disorder are endorsing the concept of partial PTSD. Journal of Traumatic Stress, 12(1), 155-165. Seedat, S., Nyamai, C., Njenga, F., Vythilingum, B. & Stein, D.J. (2004) Trauma exposure and post-traumatic stress symptoms in urban African schools. British Journal of Psychiatry, 184, 169-175. Shevlin, M. & Elklit, A. (in press). A latent class analysis of adolescent adverse life events based on a Danish national youth probability sample. Journal of Nordic Psychiatry. Tolin, D. F., & Foa, E. B. (2006). Sex differences in trauma and posttraumatic stress disorder: A quantitative review of 25 years of research. Psychological Bulletin, 132(6), 959-992. Wilson, J. (1994). The historical evolution of PTSD diagnostic criteria: From Freud to DSM-IV. Journal of Traumatic Stress, 7(4), 681-698. World Health Organization (1993). ICD-10. The ICD-10 classification of mental and behavioural disorders. ISBN 92 4 154455 4. This document is copyrighted by the American Psychological Association o r one of its allied publishers. This article is intended solely for the personal use of the individual u ser and is not to be disseminated broadly.