Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

A Longitudinal Examination of the Relationship between Trauma-Related Cognitive Factors and Internalising and Externalising Psychopathology in Physically Injured Children

A Longitudinal Examination of the Relationship between Trauma-Related Cognitive Factors and... Cognitive models of posttraumatic stress disorder (PTSD) highlight maladaptive posttrauma appraisals, trauma memory quali- ties, and coping strategies, such as rumination or thought suppression, as key processes that maintain PTSD symptoms. Anxiety, depression and externalising symptoms can also present in children in the aftermath of trauma, yet there has been little empirical investigation of the potential relevance of posttrauma cognitive processes for such difficulties. Here, we examined whether: a) acute maladaptive cognitive processes (specifically, maladaptive appraisals, memory qualities, and cognitive coping) were associated with symptoms of PTSD, internalising, and externalising at 1-month posttrauma (T1); and b) changes in these cognitive processes predicted symptom change at a follow-up assessment 6 months later (T2). We recruited 132 6–13 year old children and their parents from emergency departments following the child’s experience of an acute trauma. Children self- reported on their maladaptive appraisals, trauma-memory and cognitive coping strategies, along with symptoms of PTSD, anxiety and depression. Parents also rated children’s internalising and externalising symptoms. We found each cognitive process to be robustly associated with PTSD and non-PTSD internalising symptoms at T1, and change in each predicted change in symptoms to T2. Maladaptive appraisals and cognitive coping were unique predictors of children’s posttrauma internalising. Effects were partially retained even controlling for co-occurring PTSD symptoms. There was less evidence that trauma-specific cognitive processes were associated with externalising symptoms. Findings suggest aspects of cognitive models of PTSD are applicable to broader posttrauma psychopathology, and have implications for how we understand and target children’s posttrauma psychological adjustment. . . . . Keywords Posttraumatic stress disorder Internalising and externalising Appraisals Memory Longitudinal Following a young person’s experience of trauma, they are at PTSD symptoms (PTSS) can be common in the posttrauma risk of developing a range of poor mental health outcomes, of period, for many young people initial symptoms will naturally which the most widely studied is posttraumatic stress disorder recede without the need for formal support (Hiller et al. 2017; (PTSD; Alisic et al. 2014;Hiller etal. 2017). While elevated Le Brocque et al. 2009), with 10–20% experiencing more Electronic supplementary material The online version of this article (https://doi.org/10.1007/s10802-018-0477-8) contains supplementary material, which is available to authorized users. * Sarah L. Halligan Oxford Health NHS Foundation Trust, Cotswold House, Specialist s.l.halligan@bath.ac.uk Eating Disorder Service, Savernake Hospital, Marlborough, UK Emergency Department, Bristol Royal Hospital for Children, Department of Psychology, University of Bath, Bath, UK Bristol, UK School of Psychology and Clinical Language Sciences, University of Faculty of Health and Applied Sciences, University of West England, Reading, Reading, UK Bristol, UK Norwich Medical School, University of East Anglia, Department of Experimental Psychology, University of Oxford, Norwich, UK Oxford, UK 684 J Abnorm Child Psychol (2019) 47:683–693 chronic distress (Alisic et al. 2014; Hiller et al. 2017). McKinnon et al. 2017). Moreover, trauma-focused cognitive However, PTSD is just one potential adverse psychological behavioural therapy (tf-CBT), the recommended first-line outcome following trauma – other internalising difficulties treatment for PTSD in children (Cohen et al. 2000; Perrin (e.g., anxiety, depression) and externalising difficulties (e.g., et al. 2000), specifically targets the reduction of maladaptive attention and behaviour problems) can also be elevated in appraisals, unhelpful coping strategies, and problematic mem- trauma-exposed young people (e.g., Pine and Cohen 2002; ory qualities. Scheeringa and Zeanah 2008). While relatively robust empir- While trauma-related cognitive processes are relatively ical evidence has identified key psychological mechanisms well-established in the development and maintenance of that contribute to the maintenance of chronic PTSS (e.g., PTSS, there has been less extensive examination of their po- Ehlers and Clark 2000; Trickey et al. 2012), the role of these tential role in relation to broader posttrauma psychopathology. processes in relation to other posttrauma psychological out- This omission is important, as other internalising difficulties comes has been little studied. Such investigation is important such as anxiety and depression, as well as externalising diffi- for understanding the extent to which targeting trauma-related culties such as behaviour problems, are also all relatively com- psychological processes in intervention is likely to address mon sequelae of trauma exposure in young people, and are other adverse mental health outcomes, beyond PTSD. It also commonly comorbid with PTSD (e.g., Cénat and Derivois has theoretical implications, especially for understanding 2015; Scheeringa and Zeanah 2008; Scheeringa 2015). whether current models specific to the maintenance of PTSD Notably, trauma-related maladaptive appraisals have been may more accurately represent models of children’sbroader found to be significantly cross-sectionally associated with posttrauma mental health. posttrauma anxiety and depression following sexual assault Theories of the development and maintenance of PTSD (e.g., Mannarino and Cohen 1996); with internalising and highlight several key psychological processes that may lead externalising symptoms in a community sample of children to, or maintain, PTSS (Brewin et al. 1996; Ehlers and Clark and teens (Liu and Chen 2015); and with depression, but not 2000; Foa et al. 1989;Meiser-Stedman 2002). For example, carer-reported externalising, following maltreatment (Leeson Ehlers and Clark’s cognitive model of PTSD (Ehlers and and Nixon 2011). This is consistent with the centrality of Clark 2000) identifies three core posttrauma processes that maladaptive, dysfunctional or distorted cognitions about the maintain PTSD, particularly by contributing to a sense of cur- self, others, and/or world, to models of child depression (e.g., rent threat. First, the individual makes strong negative ap- Birmaher et al. 1996), anxiety (e.g., Ellis and Hudson 2010; praisals of the event or their own responses (e.g., BI’ll never Rapee and Heimberg 1997), and conduct problems (e.g., get over what happened^). Second, the trauma memory is Dodge and Pettit 2003). The role of posttrauma coping and proposed to be encoded in such a way that it is particularly memory quality in relation to broader psychopathology has prone to retrieval via direct triggering through matching cues been less well established. However, rumination has been as- and has ‘flashback’ like properties (e.g., due to a high level of sociated with non-trauma specific depression in children (e.g., sensory content, and limited contextual or semantic see Hitchcock et al. 2014), while avoidant coping is central to encoding). Third, engagement in maladaptive coping strate- the maintenance of a broad range of child internalising diffi- gies is hypothesised to maintain maladaptive appraisal and culties (e.g., Murray et al. 2009). Equally, researchers have trauma memory properties (e.g., due to cognitive avoidance), highlighted the presence of negative intrusive memories in and directly exacerbate symptoms (e.g., due to counterproduc- depression and in anxiety disorders (e.g., Hackmann and tive attempts to suppress trauma memories). Holmes 2004; Patel et al. 2007) The empirical literature, including a number of meta- The primary aim of the current study was to investigate the analytic reviews, provides strong support for the central role potential role of key trauma-related cognitive processes in of posttrauma cognitive processes in the maintenance of trau- relation to a range of psychological outcomes in children, in matic stress. For example, meta-analytic reviews of the child a longitudinal study of an acute-trauma exposed sample and adolescent trauma literature have concluded that the use (Hiller et al. 2017). We examined whether: a) trauma related of thought suppression (a maladaptive cognitive coping strat- appraisals, memory qualities, and maladaptive coping strate- egy) is moderately to strongly positively associated with gies showed cross-sectional associations with anxiety, depres- PTSS severity (Trickey et al. 2012), and that there is a strong sion and externalising symptoms, measured 1-month association between maladaptive appraisals and PTSS posttrauma, as well as with symptoms of PTSD; and b) wheth- (Mitchell et al. 2017; Trickey et al. 2012). Theroleof er changes in cognitive processes from 1 to 7 months were trauma-related memory quality is more well-established in associated with internalising and externalising severity across the adult PTSD field (Brewin 2014), although there is some this period. Given that there is overlap between symptoms of evidence for a significant association between sensory-laden PTSD and those of depression and anxiety problems, we also trauma memories and higher PTSS among children following examined whether posttrauma cognitive mechanisms made a trauma (Salmond et al. 2011;Meiser-Stedman et al. 2007a; unique contribution to the prediction of anxiety, depression, J Abnorm Child Psychol (2019) 47:683–693 685 and/or externalising symptoms once levels of PTSS were con- was assigned using the Manchester Triage System (nurse rat- trolled for. Although a particularly stringent analysis, this ing of urgency of care), ranging from 1 = immediate care allowed for stronger inferences in relation to whether required to 4 = standard (i.e., less urgent). Key study indices posttrauma psychological processes are relevant intervention were then obtained through child and parental questionnaires, targets even in the absence of significant PTSS. completed at both T1 and T2. Specifically, we utilised child self-report measures of cognitive processes, PTSS, anxiety and depressive symptoms. By contrast, we used parental re- Method port as our main measure of child externalising, as is recom- mended in the literature (Stanger and Lewis 1993). Participants Maladaptive Trauma Appraisals Children completed the Participants were 132 children aged 6–13 years old, and their Children’s Posttraumatic Cognitions Inventory (CPTCI), a 25- caregivers, recruited following the child’s involvement in a item self-report measure suitable for children as young as 6 years trauma and subsequent attendance at the emergency depart- old (Meiser-Stedman et al. 2009). The CPTCI measures ap- ment (ED). Participants were recruited between April 2014 praisals relating to: (i) permanent and disturbing change (e.g., and January 2016, from four EDs in the UK, as previously BMy reactions since the event mean I have changed for the reported (Hiller et al. 2017). Exclusion criteria were: signifi- worse^)and(ii) beingafragilepersonina scaryworld(e.g., cant learning difficulty or neurodevelopmental disorder; or- BAnyone could hurt me^). Young people rate their agreement ganic traumatic brain injury; suspicion that the injury was with each statement from 1 = don’t agree at all to 4 = agree a lot, caused by the young person themselves or their carer. with the total score providing an overall index of maladaptive Thirty-nine percent of all potentially eligible participants were appraisals (ranging from 25 to 100; α = 0.93). recruited to the final study (main reason for non-participation was that families could not be contacted). We found no evi- Posttrauma Coping Children completed the Child Posttrauma dence that study participants differed from the eligible popu- Coping Questionnaire (CPCQ), an 11-item self-report scale lation. Complete recruitment details are presented in Hiller created for this study based on cognitive coping items used et al. 2017, while the recruitment flow-chart is available in in previous research (Ehlers and Clark 2000;Ehlers supplementary materials (S1). Index traumas were: motor ve- et al. 2003; Stallard 2003). The measure includes 5 items hicle accident (n = 68, 52%), fall from a height (n = 25, 19%), on rumination (e.g., BIcan’t stop thinking if only the event significant bicycle accident (n = 9, 7%), acute medical episode hadn’t happened to me^) and 6 items on thought suppression (n = 10, 8%; e.g., acute anaphylaxis), sport injury (n =6, 5%), (e.g., BI’ve tried to keep any memories of what happened out and assault (n = 3, 2%) or other event (n = 17, 13%, e.g., house of my head^), each rated on a four-point scale from 0 = not at fire, dog attack, near drowning, sports injury). all or only one time to 3 = a lot of the time, and summed to yield a total score (between 0 and 30). The measure shows Procedure good internal consistency (α = 0.89) and validity against a measure of child PTSS symptoms (Hiller et al. 2017). ED staff initially approached families and obtained permission for the research team to make contact in order to confirm Trauma Memory Quality Children completed an adapted ver- eligibility and recruit them to the study. Informed consent sion of the Trauma Memory Quality Questionnaire was provided by the caregiver, while informed assent was (TMQQ; Meiser-Stedman et al. 2007a). This 18-item mea- provided by the young person. Data presented in the current sure covers the original scale’s 11-items on the sensory paper are from home assessments completed independently quality of the trauma-memory and sense of Bnowness^ by parents and children at 2–6 weeks post-ED attendance (e.g., BMy memories of the event are mostly pictures of (T1) and 6-months later (T2; i.e., 7-months posttrauma). images^; BWhen I think about the frightening event I can Dyads received GBP20 at each time point, as a thank you sometimes smell things that I smelt when the frightening for their time. An additional 3-month postal questionnaire event happened^), and an additional 7-items on assessment also took place, as reported in Hiller et al. 2017. disorganised memories (e.g., BI get mixed up about what Ninety-six percent of the sample were retained at the 6-month order things happened in during the frightening event^), assessment. adapted from an equivalent adult scale (e.g., Halligan et al. 2003). Items are rated on a scale from 1 = disagree a Measures lot to 4 = agree a lot and summedtoyieldatotal score (range, 18–72). Internal consistency of the 18-item scale Demographic and trauma-related information were obtained was α = 0.86, compared to α = 0.80 for the original 11- from ED notes and parent interview. Objective trauma severity item scale. 686 J Abnorm Child Psychol (2019) 47:683–693 PTSS Children completed the child self-report version of the controlling for PTSS. We used preliminary correlational anal- PTSD Reaction Index (PTSD-RI), which has established reli- yses to establish which cognitive variables showed significant ability and validity for children as young as 6 years old univariate associations with mental health outcomes (p <.05), (Steinberg et al. 2004). The PTSD-RI assesses 17 DSM-IV- for inclusion in subsequent regression analyses. To index T1- TR PTSD symptoms. Responses to each item are rated on a 5- T2 change in cognitions we generated residual change scores point Likert scale, ranging from 0 = none of the time to 4 = via linear regressions. Next, we ran separate linear regressions most of the time, and yielding a total symptom score as the to explore which of the three cognitive processes (measured at main outcome (ranging from 0 to 68; α =0.89). T1) most strongly predicted 1-month symptom severity (T1 PTSS, anxiety, depression, parent-reported internalising and Anxiety and Depression Children completed the 25-item short externalising), when they were examined simultaneously (i.e., version of the Revised Child Anxiety and Depression Scale entered into a single step in the regression). For non-PTSS (RCADS; Chorpitaetal. 2000; Ebesutani et al. 2012), a wide- outcomes, additional models tested for predictive effects after ly used, self-report measure, validated with children aged controlling co-occurring PTSS. Next, to explore the longitu- 7 years and over (Ebesutani et al. 2012). Items are rated on a dinal association between cognitions and symptoms, we 4-point scale ranging from 0 = never to 3 = always,measuring looked at whether change in cognitive processes was associ- 15 symptoms of anxiety (e.g., BI worry about being away from ated with change in symptoms, by running linear regression my parents^; total score range 0–45; α =0.89) and 10 depres- models with T2 symptoms as the dependent variable, whilst sive symptoms (e.g., BIfeel worthless^; total score range 0– controlling for T1 symptoms in the model. In particularly 30; α =0.86). conservative analyses, these regressions for non-PTSS out- comes were re-run also controlling for change in PTSS (resid- Parent-Report on Child Internalising and Externalising Parents ual change scores), to explore whether change in the cognitive completed the Strengths and Difficulties Questionnaire (SDQ) processes would continue to predict depression/anxiety/ parent-report version, a widely used measure of children’s internalising/externalising symptoms change, even above emotional and behavioural difficulties, suitable for reporting what is explained by PTSS. Age, sex and triage (a marker of on children as young as 4 years of age (Goodman 1997). objective trauma severity) were explored as potential Parents rate items from 0 = not true to 2 = certainly true.We covariates. utilised the 10-item externalising subscale which comprises Overall there was little missing data (maximum missing items covering hyperactivity and conduct problems (e.g., data was less than 10%). The only exception to this was the BOften has temper tantrums or hot tempers^, BEasily T1 SDQ, which was added after the study begun and was distracted^; total score range 0–20; α = 0.76); and the 10- missing for 20% of the sample. In all cases data were missing completely at random. As a sensitivity analysis, to account for item internalising scale, which assesses emotional and peer problems (e.g., BMany worries, often seems worried^; missing data, key analyses were run using multiple imputation BPicked on by other children^; total score range 0–20; α = with 50 iterations and predictive mean matching. The pattern 0.80). The SDQ was added to the T1 assessment during the of results was the same and completer-only data are presented course of the study; consequently, T1 data is only available for here. 106 participants. Data Analytic Plan Results Data were analysed using IBM SPSS Statistics for Descriptive Information Windows v22 (2013, IBM Corp., Armonk, NY). All mental- health outcome measure scores were positively skewed, so a Sample characteristics are presented in Table 1. The sample square-root transformation was applied. As most 6-month da- comprised 132 children (62% boys), aged 6–13 years old, and ta remained significantly skewed, associations were checked their participating parent (90% mothers). Child age, sex, and against non-parametric tests (Spearman’s rho), with any dis- triage category were assessed as potential covariates. The crepancies noted in Table 2. Our primary aims were to explore child’s age was negatively associated with PTSS at both time (1) whether the three cognitive processes (appraisals, memory, points (r > −0.18, p > 0.04). Sex was significantly associated coping) would be associated with acute (1-month, T1) symp- with initial anxiety (r = 0.21, p = 0.04) and depression (r = tom severity, and (2) whether change in these processes from 1 0.20, p = 0.049) symptom severity, with higher symptom to 7 months posttrauma (T1 – T2) would be associated with scores associated with being female. Triage ratings were not change in symptoms over this period. In all cases, we also significantly associated with any measures of mental health examined whether cognitive processes would continue to be (all ps > 0.13). Age and sex were controlled in all subsequent significantly associated with symptom outcomes, even after analyses. J Abnorm Child Psychol (2019) 47:683–693 687 Table 1 Descriptive information (appraisals β =0.43, p <0.001; memory β =0.15, p =0.034; coping β =0.17, p =0.018). Demographic characteristics Statistic (N =132) Parent characteristics Anxiety Symptoms Age in years (M[SD]) 39.7 (7.0) Proportion mothers 119 (90%) Each cognitive process was significantly positively correlated Proportion married or cohabiting 97 (74%) with child reported anxiety at T1, and change in each process was associated with equivalent change in anxiety scores from Education status: School until 16 years or younger 36 (27%) Further education 50 (38%) T1 to T2 (see Tables 2 and 3). In a regression model control- ling for age and sex, combined T1 cognitive processes ex- Higher education 46 (35%) plained 63% of variance in T1 anxiety scores, with each being Child characteristics a unique predictor (see Table 4). When additionally control- Age in years, M(SD) 9.87 (1.8) ling for T1 PTSS, the combined cognitive processes still ex- Male 82 (62.1%) plained 12% of variance in T1 anxiety, and maladaptive ap- Ethnicity – Caucasian 121 (91.7%) praisals and coping remained unique predictors (see Table 4). Triage category We next tested whether change in cognitive processes sig- 1 (immediate attention required) 61 (46%) nificantly predicted change in anxiety from T1 to T2, by con- 2 (very urgent) 29 (22%) trolling for T1 anxiety scores (and age and sex) in a regression 3 (urgent) 26 (20%) model with T2 anxiety as the dependent variable. In combina- 4 (less urgent) 16 (12%) tion, cognitive change scores explained 28% of variance in this Days in hospital (Min – Max, M [SD]) 0–28, 2.64 (4.83) a model, with change in maladaptive appraisals and cognitive Days of school missed (Min – Max, M [SD]) 0–28, 5.52 (5.98) coping, but not trauma-memory, uniquely predicting change Proportion requiring ambulance/helicopter 90 (70%) in anxiety symptoms across the 6-months (see Table 5). Proportion with head injury 33 (25%) When additionally controlling for change in PTSS, combined Days of school missed represents days missed prior to their first cognitive processes remained a significant predictor of anxiety, assessment explaining 3% of variance, although no individual cognitive processes uniquely predicted anxiety change (Table 5). Paired samples t-tests indicated that children’s self-reported symptoms significantly reduced from 1-month to 6-months Depressive Symptoms across all measures (M and SDs presented in Table 2; PTSS, p < 0.001; Anxiety, p< 0.001; Depression, p < 0.001). By con- Each cognitive process showed significant positive associa- trast, there was no significant change in parent-report of child tions with T1 depression and with change in depression from internalising (p = 0.75) or externalising symptoms (p = 0.47). T1 to T2 (see Tables 2 and 3). In a regression analysis, com- bined T1 cognitive processes explained 59% of variance in T1 depression, F(3,114) = 62.64, p < 0.001, with all three pro- Posttraumatic Stress Symptoms cesses being significant independent predictors (see Table 4). When additionally controlling for T1 PTSS, cognitive pro- Significant bivariate correlations were present between all cesses explained 5% of the variance in depression scores, cognitive processes (i.e., appraisals, trauma memory, cogni- and maladaptive appraisals remained uniquely associated with tive coping) and PTSS at T1, and change in each process was depression severity (Table 4). significantly associated with change in PTSS scores from T1 Next, after controlling for initial depression severity (and to T2 (see correlation matrices in Tables 2 and 3). In a regres- age and sex), change in cognitions from T1 to T2 significantly sion model controlling for age and sex, the combined T1 cog- predicted T2 depression scores, explaining 28% of variance. nitive processes explained 72% of variance in T1 PTSS, Here change in maladaptive appraisals and cognitive coping F(3,114) = 115.98, p < 0.001, with appraisals (β = 0.30, both uniquely predicted depression severity. However, after p < 0.001), memory (β =0.33, p < 0.001) and coping (β = additionally controlling for change in PTSS scores, change 0.41, p < 0.001) each being unique predictors in the model. in cognitive processes no longer significantly predicted We next tested whether change in cognitive processes predict- changes in depression (see Table 5). ed change in PTSS, by examining prediction of T2 PTSS by cognitive change scores, in a regression model controlling for Parent-Reported Child Internalising Symptoms T1 PTSS (as well as age, sex). Combined cognitive processes explained 41% of variance, F(3,102) = 43.11, p <0.001, and Each cognitive process was correlated with parent-reported each remained a unique predictor in this longitudinal model child internalising scores at T1, but only maladaptive 688 J Abnorm Child Psychol (2019) 47:683–693 Table 2 Bivariate correlation matrix for associations between study variables 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 1-mo Processes 1. Appraisals 2. Coping 0.61** 3. Memory 0.59** 0.63** 1-mo Outcomes 4. PTSS 0.68** 0.79** 0.75** 5. Anxiety 0.74** 0.68** 0.61** 0.74** 6. Depression 0.70** 0.69** 0.58** 0.79** 0.84** a + 7. Internalising 0.37** 0.23* 0.22* 0.25** 0.18 0.24* 8. Externalising 0.29** 0.09 0.03 0.21 0.24* 0.29** 0.51** 6-mo Outcomes 9. PTSS 0.59** 0.54** 0.39** 0.62** 0.68** 0.68** 0.21* 0.30** 10. Anxiety 0.62** 0.43** 0.35** 0.54** 0.67** 0.64** 0.12 0.29** 0.75** 11. Depression 0.61** 0.45** 0.30** 0.54** 0.67** 0.71** 0.25* 0.34** 0.79** 0.84** a + 12. Internalising 0.28** 0.13 0.17 0.29** 0.36** 0.41** 0.53** 0.55** 0.49** 0.39** 0.55** 13. Externalising 0.23* −0.003 0.06 0.13 0.26* 0.21 0.33** 0.61** 0.29** 0.27* 0.39** 0.67** M (SD) 40.20 14.52 37.51 18.84 9.65 6.30 3.84 6.01 12.86 6.67 4.08 3.72 6.27 (13.47) (8.86) (10.31) (13.54) (8.40) (6.67) (3.71) (3.89) (11.88) (7.18) (4.82) (4.05) (4.79) Using Spearman’s rho, 1-mo parent-report internalising was no longer significantly correlated with coping (r =0.20, p =0.06) but 6-mo parent- reported internalising was significantly correlated with coping (r =0.26, p = 0.02). Coping was retained in the regression models. There were no other discrepancies between parametric (Bivariate) and non-parametric (Spearman’sRho) tests p < 0.10, *p< 0.05, **p < 0.001. PTSS = child-reported posttraumatic stress symptom severity. ‘Internalising’ and ‘Externalising’ based on parent- report on the SDQ, all other questionnaires were child self-report. Possible ranges of scores for each measure are presented in the Method appraisals and coping (not memory) were significantly asso- As only change in appraisals and coping from T1 to ciated with change in symptoms from T1 to T2 (see Table 3). T2 were significantly associated with change in At T1, in combination, the acute cognitive processes ex- internalising severity (Table 3), only these processes plained 15% of variance in parent-reported internalising, with were examined further in regression models. In a model maladaptive appraisals being a significant unique predictor controlling for initial internalising scores, cognitive (see Table 4). After additionally controlling for T1 PTSS, change scores in combination explained a significant cognitive processes no longer explained significant variance 10% of variance in 6-month internalising problems. in T1 parent-report internalising, although maladaptive ap- Here, change in the child’s cognitive coping was praisals remained a unique significant predictor (Table 4). uniquely associated with change in internalising Table 3 Bivariate correlations between residual change scores 1. 2. 3. 4. 5. 6. 7. 1. Appraisals 2. Coping 0.59** 3. Memory 0.55** 0.50** 4. PTSS 0.77** 0.56** 0.56** 5. Anxiety 0.63** 0.51** 0.41** 0.67** 6. Depression 0.61** 0.47** 0.36** 0.67** 0.72** 7. Internalising 0.25* 0.33** 0.16 0.28** 0.31** 0.40** 8. Externalising 0.22* 0.19 0.08 0.12 0.13 0.27** 0.52** Using Spearman’s Rho, change in appraisals was not significantly associated with change in externalising (r= 0.12, p = 0.29). There were no other discrepancies between parametric (Bivariate) and non-parametric (Spearman’sRho) tests *p <0.05, ** p < 0.01. All scores are 1–6 month residual change scores. Internalising and Externalising based on parent report, all other scores based on child self-report J Abnorm Child Psychol (2019) 47:683–693 689 Table 4 Results of linear regressions for T1 cognitive predictors of T1 Table 5 Results of linear regressions analyses of longitudinal data symptoms, after controlling for age and sex (Model 1) and T1 PTSS examining change in cognitive processes as a predictor of symptom (Model 2) change (Model 1), including when controlling for change in PTSS (Model 2) Model 1 Model 2 Model 1 Model 2 Child Report 2 2 Child Report Anxiety R =0.63, R =0.12, 2 2 F(3,114) = 77.6** F(3,113) = 14.7** Anxiety R =0.28, R =0.03, F(3,102) = 22.4** F(3,101) = 3.0* Appraisals 0.51** 0.45** Appraisals 0.38** 0.19 Memory 0.16* 0.10 Memory 0.02 −0.03 Coping 0.28** 0.21** 2 2 Coping 0.20* 0.15 Depression R =0.59, R =0.05, 2 2 F(3,114) = 62.6*** F(3,113) = 6.04*** Depression R =0.28, R =0.03, F(3,102) = 23.3*** F(3,101) = 2.4 Appraisals 0.44** 0.29** Appraisals 0.41** 0.17 Memory 0.19* 0.06 Memory 0.003 −0.06 Coping 0.30** 0.15 Coping 0.19* 0.12 Parent Report 2 2 + Parent Report Internalising R =0.15, R =0.06, F(3,91) = 2.2 2 2 F(3,92) = 5.4** Internalising R =0.10, F(2,81) = 6.6** R =0.03, F(2,80) = 2.2 Appraisals 0.36** 0.34* Appraisals 0.03 −0.16 Memory 0.06 0.03 Coping 0.30* 0.24* 2 + 2 Coping −.01 −0.05 Externalising R =0.02,F(1,82) = 3.4 R =0.01,F(1,81) = 2.0 2 2 Externalising R =0.07, R =0.07,F(1,93) = 4.2 Appraisals 0.16 0.19 F(1,94) = 7.5** Appraisals 0.28** 0.28* R and F statistics relate to the change in model fit following the inclusion of cognitive variables. For the change models, we examined whether T1 R and F statistics relate to the change in model fit following the inclusion to T2 residualized change scores for cognitive processes predicted T2 of cognitive variables, controlling for age and sex of the child (Model 1) symptom severity, controlling for age, sex and T1 symptoms (Model 1), and additionally for PTSS (Model 2) and then once also controlling for residualized change in PTSS (Model 2) ** p ≤ 0.01, *p<0.05, p < 0.10 ** p ≤ 0.01, *p<0.05, p < 0.10 severity, but appraisal change was not (Table 5). After controlling for change in PTSS, combined cognitive Discussion processes no longer explained significant variance, al- though change in cognitive coping remained a unique We examined trauma-related psychological processes (ap- significant predictor of change in parent-report praisals, coping, memory) as potential predictors of children’s internalising (Table 5). broader psychological outcomes following exposure to acute trauma. We replicated findings that all three processes explain unique variance in acute PTSS and reductions in PTSS over Parent-Reported Child Externalising Symptoms time. We also found robust evidence that children’s posttrauma cognitive processes were cross-sectionally and Only maladaptive appraisals showed significant bivariate longitudinally associated with non-PTSD internalising prob- lems, based on both self- and parent-report. These associations associations with externalising symptoms at either T1 or with change at T2, and was considered in regression with wider child internalising symptoms were particularly re- models (see Tables 2 and 3). After controlling for age liable for child-reported maladaptive appraisals and maladap- and sex, initial maladaptive appraisal scores were signif- tive cognitive coping strategies (i.e., thought suppression and icantly associated with parent-reported externalising rumination); and were partially retained even after levels of symptoms at T1, explaining 7% of variance. This effect PTSS were controlled for. There was less evidence for the role remained significant after controlling for concurrent 1- of trauma related cognitive processes in predicting externalising difficulties. month PTSS, explaining 4% of variance in parent- reported child externalising symptoms (see Table 4). Our findings add to the limited literature showing trauma- related cognitive processes are important for children’s However, after controlling for initial externalising symp- toms (and age and sex), change in maladaptive ap- broader mental health following trauma. In particular, it builds on cross-sectional studies that have demonstrated associations praisals from T1 to T2 was not significantly predictive of T2 externalising (Table 5). between trauma-related negative appraisals and broader 690 J Abnorm Child Psychol (2019) 47:683–693 psychopathology (e.g., Leeson and Nixon 2011; Liu and Chen was no evidence that memory quality predicted wider 2015). The current study found that maladaptive appraisals, internalising problems once PTSS was controlled for. Caution coping and trauma memory quality at 1-month posttrauma is warranted when interpreting these null findings, due to the were each moderately to strongly correlated with children’s strong overlap between all three processes and sets of symp- self-reported depression and anxiety symptoms, and showed toms. Nonetheless, in this case we cannot rule out the possibil- slightly smaller associations with parent-reported child ity that associations with anxiety and depression are accounted internalising symptoms. Each cognitive process explained for by the co-occurrence of PTSD symptomatology. Whereas unique variance in child reported non-PTSD internalising negative appraisals and maladaptive coping strategies are wide- symptoms at 1-month posttrauma, whereas negative ap- ly implicated in models of anxiety (e.g., thought suppression) praisals were the only independent predictor of parent- and depression (rumination), trauma-memory quality may be a reported internalising at this timepoint. Longitudinal analyses unique driver of traumatic stress. This is consistent with theo- examining change in cognitive processes over time found a retical perspectives that highlight traumatic memories as a key similar pattern of results. Together, changes in maladaptive defining feature of PTSD versus other emotional disorders, cognitive processes explained a relatively large 28% of vari- both phenomenologically and in terms of underlying neural ance in change in anxiety and depression across a 6-month processes (e.g., Elzinga and Bremner 2002). follow-up, and a smaller 10% of variance in change in parent- Overall, our evidence highlights the potential utility of con- reported internalising problems. Changes in appraisals and sidering trauma-related maladaptive coping strategies and ap- coping were identified as uniquely associated with symptom praisals as potential treatment targets in depressed or anxious change when cognitive processes were examined simulta- youth who present following trauma exposure. That the same neously as predictors. processes that maintain PTSS also contribute to non-PTSD Although not the main focus of the current study, we also internalising may also explain why many CBT trials targeting found robust longitudinal evidence that appraisals, memory PTSD see concurrent reductions in internalising comorbidities and maladaptive coping strategies each explain unique variance (e.g., Deblinger et al. 2011; Goldbeck et al. 2016; Smith et al. in children’s PTSS, consistent with cognitive models (e.g., 2007). Our findings suggest that this change in broader psycho- Ehlers and Clark 2000). In particularly stringent analyses, we pathology likely particularly results from the targeting of chil- examined whether cognitive processes could predict broader dren’s maladaptive appraisals and coping strategies. While our internalising symptoms when controlling for co-occurring findings suggest that a robust focus on appraisals and maladap- PTSS. This substantially reduced the variance explained by tive coping may be particularly clinically useful when young cognitive factors, which is unsurprising given the strong corre- people present with internalising problems linked to trauma, lation and symptom overlap between PTSS, anxiety and de- replication is required before drawing clinical conclusions. pression. Nevertheless, even after controlling for concurrent We found substantially less evidence that posttrauma cog- PTSS we still found that maladaptive appraisals (e.g., Bthe nitive processes contribute to child externalising symptoms. world is unsafe^, BI’ll never get over what happened^)and Our findings build on Liu and Chen’s(2015) study with a cognitive coping (i.e., thought suppression, rumination), community-sample of adolescents, replicating a cross- remained cross-sectionally associated with children’s non- sectional association between appraisals and externalising, PTSD internalising at 1-month posttrauma. Moreover, in lon- but additionally suggest that changes in trauma-related ap- gitudinal analyses, reductions in maladaptive cognitive coping praisals do not drive longer-term adjustment in terms of over time continued uniquely to predict change in parent- externalising behaviour. We found no evidence that trauma reported child internalising problems, while the combined cog- memory quality or maladaptive cognitive coping relate to chil- nitive processes significantly predicted reductions in child re- dren’s posttrauma externalising symptoms. The suggestion ported anxiety. This indicates that associations are unlikely to that these psychological processes may be less relevant to be secondary to PTSD symptoms, but rather that cognitive the maintenance of child externalising is consistent with some processes potentially make a direct contribution to children’s findings from treatment studies. For example, Deblinger et al. wider internalising problems following trauma. (2011) found that externalising was more effectively targeted It is notable that whereas negative appraisals and maladap- in a treatment that had a larger focus on parent-training, com- tive coping strategies were relatively consistently associated pared to a treatment that placed more focus on children’s with children’s anxiety and depression, there was less robust trauma-related memories and appraisals. Thus, characteristics evidence for the role of trauma-related memory qualities in of children’s posttrauma social environment may be more rel- relation to longer-term broader psychopathology. Thus, al- evant to managing posttrauma externalising behaviours though trauma memory quality showed a clear pattern of bivar- (Deblinger et al. 1996; Silverman et al. 2008). iate associations with child internalising symptoms, it did not Findings should be interpreted in the light of limitations. emerge reliably as a unique predictor of distress over and above First, as this is an observational longitudinal study, causation other processes, particularly in longitudinal models; and there cannot be determined. Second, we did not include a child- J Abnorm Child Psychol (2019) 47:683–693 691 and RMS is an NIHR Career Development Fellow (CDF-2015-08-073). report measure of externalising symptoms in our study, and we The views expressed are those of the authors and not necessarily those of were unable to compare exact measures of internalising (i.e., the NHS, the NIHR or the Department of Health. anxiety and depression) across child- versus parent-report. The approach we took to symptom measurement is consistent with Funding This research was funded by an Economic and Social Research best practice: parent or teacher reports are considered the gold Council (ESRC) grant awarded to SLH (ES/K006290/1). standard for measuring children’s externalising (Stanger and Lewis 1993), whereas child-report is recommended for mea- Compliance with Ethical Standards surement of PTSS (e.g., Kassam-Adams et al. 2006;Meiser- Conflicts of Interest None. Stedman et al. 2007b). Our observation that posttrauma child internalising problems reduced from 1-month to 6-months ac- Ethical Approval Ethical approval was obtained from the University of cording to child-report but not parent-report may suggest that Bath Research Ethics Committee and the NHS Research Ethics parents are reporting on children’s pre-established psychologi- Committee South Central - Oxford A (Ref 137454). cal profiles, rather than detecting changes in their posttrauma mental health. This is consistent with the view that parents may Data Access Halligan, Sarah and Hiller, Rachel (2017). The role of trauma specific behaviours and parenting style in facilitating child psy- be relatively poor at detecting children’s posttrauma symptoms. chological adjustment. [Data Collection]. Colchester, Essex: UK Data Parental perceptions of their child’s distress may also be influ- Archive https://doi.org/10.5255/UKDA-SN-852668 enced by their own posttrauma distress. Nonetheless, single- Open Access This article is distributed under the terms of the Creative informant bias may substantially inflate associations in the con- Commons Attribution 4.0 International License (http:// text of psychopathology, and associations between child- creativecommons.org/licenses/by/4.0/), which permits unrestricted use, reported symptoms and cognitions should be considered with distribution, and reproduction in any medium, provided you give this in mind. Null findings should also be considered in the appropriate credit to the original author(s) and the source, provide a link light of the relatively modest overall sample size. The sample to the Creative Commons license, and indicate if changes were made. size also meant we were not appropriately powered to run more sophisticated analyses, such as structural equation models, which may be better able to account for covariance between References the different symptom outcomes. Finally, it is notable that, as is typical of low-risk acute-incident emergency department sam- Alisic, E., Zalta, A. K., Van Wesel, F., Larsen, S. E., Hafstad, G. S., ples, overall internalising and externalising symptoms in our Hassanpour, K., & Smid, G. E. (2014). Rates of posttraumatic stress sample were relatively low. For the majority of participants the disorder in trauma-exposed children and adolescents: Meta-analysis. The British Journal of Psychiatry, 204,335–340. index trauma was an acute accidental injury, and while families Birmaher, B., Ryan, N. D., Williamson, D. E., Brent, D. A., Kaufman, J., were from a range of socio-economic backgrounds, the sample Dahl, R. E., et al. (1996). Childhood and adolescent depression: A was mostly culturally- and ethnically-homogenous, and cap- review of the past 10 years. Part I. Journal of the American tured children up to 13 years old. Findings cannot be general- Academy of Child & Adolescent Psychiatry, 35(11), 1427–1439. ised to more complex traumas or high-risk groups, or to ado- Brewin, C. R. (2014). Episodic memory, perceptual memory, and their interaction: Foundations for a theory of posttraumatic stress disor- lescents or other cultural contexts. der. Psychological Bulletin, 140(1), 69. In conclusion, cognitive processes that are known to be Brewin, C. R., Dalgleish, T., & Joseph, S. (1996). A dual representation important in the maintenance of PTSD are also likely to be theory of posttraumatic stress disorder. Psychological Review, relevant for children’s broader posttrauma internalising diffi- 103(4), 670–686. culties. Our findings particularly highlight the role of mal- Cénat, J. M., & Derivois, D. (2015). Long-term outcomes among child and adolescent survivors of the 2010 Haitian earthquake. adaptive coping and maladaptive appraisals, in predicting Depression and Anxiety, 32(1), 57–63. acute distress and reductions in non-PTSD internalising prob- Chorpita, B. F., Yim, L., Moffitt, C., Umemoto, L. A., & Francis, S. E. lems, and suggest that it may be beneficial to take a trauma- (2000). Assessment of symptoms of DSM-IV anxiety and depres- informed approach to addressing posttrauma internalising sion in children: A revised child anxiety and depression scale. problems by targeting these processes. What remains to be Behaviour Research and Therapy, 38(8), 835–855. Cohen, J. A., Mannarino, A. P., Berliner, L., & Deblinger, E. (2000). seen is whether trauma informed treatment frameworks may Trauma-focused cognitive behavioral therapy for children and ado- effectively target trauma-related anxiety and depression in the lescents: An empirical update. Journal of Interpersonal Violence, absence of clinical-level PTSD. 15(11), 1202–1223. Deblinger, E., Lippman, J., & Steer, R. (1996). Sexually abused children Acknowledgements We thank families for their generosity in participat- suffering posttraumatic stress symptoms: Initial treatment outcome ing; clinical teams at the Bristol Royal Hospital for Children, Royal findings. Child Maltreatment, 1,310–321. United Hospital Bath, Great Western Hospital, and the Gloucestershire Deblinger, E., Mannarino, A. P., Cohen, J. A., Runyon, M. K., & Steer, R. Royal Hospital for supporting our research; and Hope Christie, Claire A. (2011). Trauma-focused cognitive behavioral therapy for chil- Dixon and Victoria Williamson for their assistance with this research. dren: Impact of the trauma narrative and treatment length. CC is funded by an NIHR Research Professorship (RP_2014-04-018) Depression and Anxiety, 28(1), 67–75. 692 J Abnorm Child Psychol (2019) 47:683–693 Dodge, K. A., & Pettit, G. S. (2003). A biopsychosocial model of the psychopathology in Taiwanese children and adolescents. Journal of Abnormal Child Psychology, 43(8), 1475–1484. development of chronic conduct problems in adolescence. Developmental Psychology, 39(2), 349. Mannarino, A. P., & Cohen, J. A. (1996). Abuse-related attributions and Ebesutani, C., Reise, S. P., Chorpita, B. F., Ale, C., Regan, J., Young, J., perceptions, general attributions, and locus of control in sexually Higa-McMillan, C., & Weisz, J. R. (2012). The revised child anxiety abused girls. Journal of Interpersonal Violence, 11(2), 162–180. and depression scale-short version: Scale reduction via exploratory McKinnon, A., Brewer, N., Cameron, K., & Nixon, R. D. (2017). The bifactor modeling of the broad anxiety factor. Psychological relationship between processing style, trauma memory processes, Assessment, 24(4), 833. and the development of posttraumatic stress symptoms in children and adolescents. Journal of Behavior Therapy and Experimental Ehlers, A., & Clark, D. M. (2000). A cognitive model of posttraumatic Psychiatry, 57,135–142. stress disorder. Behaviour Research and Therapy, 38,319–345. Meiser-Stedman, R. (2002). Towards a cognitive–behavioral model of Ehlers, A., Mayou, R. A., & Bryant, B. (2003). Cognitive predictors of PTSD in children and adolescents. Clinical Child and Family posttraumatic stress disorder in children: Results of a prospective Psychology Review, 5(4), 217–232. longitudinal study. Behaviour Research and Therapy, 41(1), 1–10. Meiser-Stedman, R., Dalgleish, T., Smith, P., Yule, W., & Glucksman, E. Ellis, D. M., & Hudson, J. L. (2010). The metacognitive model of gen- (2007a). Diagnostic, demographic, memory quality, and cognitive eralized anxiety disorder in children and adolescents. Clinical Child variables associated with acute stress disorder in children and ado- and Family Psychology Review, 13(2), 151–163. lescents. Journal of Abnormal Psychology, 116(1), 65–79. Elzinga, B. M., & Bremner, J. D. (2002). Are the neural substrates of Meiser-Stedman, R., Smith, P., Glucksman, E., Yule, W., & Dalgleish, T. memory the final common pathway in posttraumatic stress disorder (2007b). Parent and child agreement for acute stress disorder, post- (PTSD)? Journal of Affective Disorders, 70(1), 1–17. traumatic stress disorder and other psychopathology in a prospective Foa, E. B., Steketee, G., & Rothbaum, B. O. (1989). Behavioral/cognitive study of children and adolescents exposed to single-event trauma. conceptualizations of post-traumatic stress disorder. Behavior Journal of Abnormal Child Psychology, 35(2), 191–201. Therapy, 20,155–176. Meiser-Stedman, R., Smith, P., Bryant, R., Salmon, K., Yule, W., Goldbeck, L., Muche, R., Sachser, C., Tutus, D., & Rosner, R. (2016). Dalgleish, T., & Nixon, R. D. (2009). Development and validation Effectiveness of trauma-focused cognitive behavioral therapy for of the child post-traumatic cognitions inventory (CPTCI). Journal of children and adolescents: A randomized controlled trial in eight Child Psychology and Psychiatry, 50(4), 432–440. German mental health clinics. Psychotherapy and Psychosomatics, Mitchell, R., Brennan, K., Curran, D., Hanna, D., & Dyer, K. F. (2017). A 85(3), 159–170. meta-analysis of the association between appraisals of trauma and Goodman, R. (1997). The strengths and difficulties questionnaire: A re- posttraumatic stress in children and adolescents. Journal of search note. Journal of Child Psychology and Psychiatry, 38,581–586. Traumatic Stress, 30(1), 88–93. Hackmann, A., & Holmes, E. (2004). Reflecting on imagery: A clinical Murray, L., Creswell, C., & Cooper, P. J. (2009). The development of perspective and overview of the special issue of memory on mental anxiety disorders in childhood: An integrative review. imagery and memory in psychopathology. Memory, 12(4), 389–402. Psychological Medicine, 39(9), 1413–1423. Halligan,S.L.,Michael,T.,Clark,D.M.,&Ehlers,A.(2003). Patel, T., Brewin, C. R., Wheatley, J., Wells, A., Fisher, P., & Myers, S. Posttraumatic stress disorder following assault: The role of cognitive (2007). Intrusive images and memories in major depression. processing, trauma memory, and appraisals. Journal of Consulting Behaviour Research and Therapy, 45(11), 2573–2580. and Clinical Psychology, 71(3), 419. Perrin, S., Smith, P., & Yule, W. (2000). Practitioner review: The assess- Hiller, R. M., Meiser-Stedman, R., Fearon, P., Lobo, S., McKinnon, A., ment and treatment of posttraumatic stress disorder in children and Fraser, A., & Halligan, S. L. (2016). Research review: Changes in adolescents. The Journal of Child Psychology and Psychiatry and the prevalence and symptom severity of child post-traumatic stress Allied Disciplines, 41(3), 277–289. disorder in the year following trauma–a meta-analytic study. Journal Pine, D. S., & Cohen, J. A. (2002). Trauma in children and adolescents: of Child Psychology and Psychiatry, 57,884–898. Risk and treatment of psychiatric sequelae. Biological Psychiatry, Hiller, R. M., Meiser-Stedman, R., Lobo, S., Creswell, C., Fearon, P., 51(7), 519–531. Ehlers, A., Murray, L., Halligan, S. L. (2017). The role of parenting Rapee, R. M., & Heimberg, R. G. (1997). A cognitive-behavioral model style and behaviours in facilitating child posttrauma psychological of anxiety in social phobia. Behaviour Research and Therapy, 35(8), adjustment. Journal of Child Psychology and Psychiatry. In Press. 741–756. Hitchcock, C., Nixon, R. D., & Weber, N. (2014). A review of Salmond, C. H., Meiser-Stedman, R., Glucksman, E., Thompson, P., overgeneral memory in child psychopathology. British Journal of Dalgleish, T., & Smith, P. (2011). The nature of trauma memories Clinical Psychology, 53(2), 170–193. in acute stress disorder in children and adolescents. Journal of Child Kassam-Adams, N., Garcia-Espana, J. F., Miller, V. A., & Winston, F. Psychology and Psychiatry, 52(5), 560–570. (2006). Parent-child agreement regarding children_s acute stress: Scheeringa,M. S.,&Zeanah,C.H.(2008).Reconsiderationof The role of parent acute stress reactions. Journal of the American harm’s way: Onsets and comorbidity patterns of disorders in Academy of Child and Adolescent Psychiatry, 45(12), 1485–1493. preschool children and their caregivers following hurricane Le Brocque, R. M., Hendrikz, J., & Kenardy, J. A. (2009). The course of Katrina. Journal of Clinical Child and Adolescent posttraumatic stress in children: Examination of recovery trajecto- Psychology, 37,508–518. ries following traumatic injury. Journal of Pediatric Psychology, Scheeringa, M. S. (2015). Untangling psychiatric comorbidity in young 35(6), 637–645. children Who experienced single, repeated, or Hurricane Katrina Leeson, F. J., & Nixon, R. D. (2011). The role of children’s appraisals on traumatic events. Child Youth Care Forum, 44, 475. https://doi. adjustment following psychological maltreatment: A pilot study. org/10.1007/s10566-014-9293-7. Journal of Abnormal Child Psychology, 39(5), 759–771. Silverman, W. K., Ortiz, C. D., Viswesvaran, C., Burns, B. J., Kolko, D. Liu, S. T., & Chen, S. H. (2015). A community study on the relationship J., Putnam, F. W., & Amaya-Jackson, L. (2008). Evidence-based of posttraumatic cognitions to internalizing and externalizing psychosocial treatments for children and adolescents exposed to J Abnorm Child Psychol (2019) 47:683–693 693 traumatic events. JournalofClinicalChild &Adolescent Stanger, C., & Lewis, M. (1993). Agreement among parents, teachers, and children on internalizing and externalizing behavior problems. Psychology, 37(1), 156–183. Smith, P., Yule, W., Perrin, S., Tranah, T., Dalgleish, T., & Clark, D. M. Journal of Clinical Child Psychology, 22(1),107–116. (2007). Cognitive-behavioral therapy for PTSD in children and ad- Steinberg, A. M., Brymer, M. J., Decker, K. B., & Pynoos, R. S. (2004). olescents: A preliminary randomized controlled trial. Journalofthe The University of California at Los Angeles post-traumatic stress American Academy of Child & Adolescent Psychiatry, 46,1051– disorder reaction index. Current Psychiatry Reports, 6(2),96–100. Trickey, D., Siddaway, A. P., Meiser-Stedman, R., Serpell, L., & Field, A. Stallard, P. (2003). A retrospective analysis to explore the applicability of P. (2012). A meta-analysis of risk factors for posttraumatic stress the Ehlers And Clark (2000) cognitive model to explain PTSD in disorder in children and adolescents. Clinical Psychology Review, children. Behavioural and Cognitive Psychotherapy, 31(3),337– 32,122–138. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Abnormal Child Psychology Springer Journals

A Longitudinal Examination of the Relationship between Trauma-Related Cognitive Factors and Internalising and Externalising Psychopathology in Physically Injured Children

Download PDF
11 pages

Loading next page...
 
/lp/springer-journals/a-longitudinal-examination-of-the-relationship-between-trauma-related-Dze72mctLQ

References (53)

Publisher
Springer Journals
Copyright
Copyright © 2018 by The Author(s)
Subject
Psychology; Child and School Psychology; Neurosciences; Public Health
ISSN
0091-0627
eISSN
1573-2835
DOI
10.1007/s10802-018-0477-8
Publisher site
See Article on Publisher Site

Abstract

Cognitive models of posttraumatic stress disorder (PTSD) highlight maladaptive posttrauma appraisals, trauma memory quali- ties, and coping strategies, such as rumination or thought suppression, as key processes that maintain PTSD symptoms. Anxiety, depression and externalising symptoms can also present in children in the aftermath of trauma, yet there has been little empirical investigation of the potential relevance of posttrauma cognitive processes for such difficulties. Here, we examined whether: a) acute maladaptive cognitive processes (specifically, maladaptive appraisals, memory qualities, and cognitive coping) were associated with symptoms of PTSD, internalising, and externalising at 1-month posttrauma (T1); and b) changes in these cognitive processes predicted symptom change at a follow-up assessment 6 months later (T2). We recruited 132 6–13 year old children and their parents from emergency departments following the child’s experience of an acute trauma. Children self- reported on their maladaptive appraisals, trauma-memory and cognitive coping strategies, along with symptoms of PTSD, anxiety and depression. Parents also rated children’s internalising and externalising symptoms. We found each cognitive process to be robustly associated with PTSD and non-PTSD internalising symptoms at T1, and change in each predicted change in symptoms to T2. Maladaptive appraisals and cognitive coping were unique predictors of children’s posttrauma internalising. Effects were partially retained even controlling for co-occurring PTSD symptoms. There was less evidence that trauma-specific cognitive processes were associated with externalising symptoms. Findings suggest aspects of cognitive models of PTSD are applicable to broader posttrauma psychopathology, and have implications for how we understand and target children’s posttrauma psychological adjustment. . . . . Keywords Posttraumatic stress disorder Internalising and externalising Appraisals Memory Longitudinal Following a young person’s experience of trauma, they are at PTSD symptoms (PTSS) can be common in the posttrauma risk of developing a range of poor mental health outcomes, of period, for many young people initial symptoms will naturally which the most widely studied is posttraumatic stress disorder recede without the need for formal support (Hiller et al. 2017; (PTSD; Alisic et al. 2014;Hiller etal. 2017). While elevated Le Brocque et al. 2009), with 10–20% experiencing more Electronic supplementary material The online version of this article (https://doi.org/10.1007/s10802-018-0477-8) contains supplementary material, which is available to authorized users. * Sarah L. Halligan Oxford Health NHS Foundation Trust, Cotswold House, Specialist s.l.halligan@bath.ac.uk Eating Disorder Service, Savernake Hospital, Marlborough, UK Emergency Department, Bristol Royal Hospital for Children, Department of Psychology, University of Bath, Bath, UK Bristol, UK School of Psychology and Clinical Language Sciences, University of Faculty of Health and Applied Sciences, University of West England, Reading, Reading, UK Bristol, UK Norwich Medical School, University of East Anglia, Department of Experimental Psychology, University of Oxford, Norwich, UK Oxford, UK 684 J Abnorm Child Psychol (2019) 47:683–693 chronic distress (Alisic et al. 2014; Hiller et al. 2017). McKinnon et al. 2017). Moreover, trauma-focused cognitive However, PTSD is just one potential adverse psychological behavioural therapy (tf-CBT), the recommended first-line outcome following trauma – other internalising difficulties treatment for PTSD in children (Cohen et al. 2000; Perrin (e.g., anxiety, depression) and externalising difficulties (e.g., et al. 2000), specifically targets the reduction of maladaptive attention and behaviour problems) can also be elevated in appraisals, unhelpful coping strategies, and problematic mem- trauma-exposed young people (e.g., Pine and Cohen 2002; ory qualities. Scheeringa and Zeanah 2008). While relatively robust empir- While trauma-related cognitive processes are relatively ical evidence has identified key psychological mechanisms well-established in the development and maintenance of that contribute to the maintenance of chronic PTSS (e.g., PTSS, there has been less extensive examination of their po- Ehlers and Clark 2000; Trickey et al. 2012), the role of these tential role in relation to broader posttrauma psychopathology. processes in relation to other posttrauma psychological out- This omission is important, as other internalising difficulties comes has been little studied. Such investigation is important such as anxiety and depression, as well as externalising diffi- for understanding the extent to which targeting trauma-related culties such as behaviour problems, are also all relatively com- psychological processes in intervention is likely to address mon sequelae of trauma exposure in young people, and are other adverse mental health outcomes, beyond PTSD. It also commonly comorbid with PTSD (e.g., Cénat and Derivois has theoretical implications, especially for understanding 2015; Scheeringa and Zeanah 2008; Scheeringa 2015). whether current models specific to the maintenance of PTSD Notably, trauma-related maladaptive appraisals have been may more accurately represent models of children’sbroader found to be significantly cross-sectionally associated with posttrauma mental health. posttrauma anxiety and depression following sexual assault Theories of the development and maintenance of PTSD (e.g., Mannarino and Cohen 1996); with internalising and highlight several key psychological processes that may lead externalising symptoms in a community sample of children to, or maintain, PTSS (Brewin et al. 1996; Ehlers and Clark and teens (Liu and Chen 2015); and with depression, but not 2000; Foa et al. 1989;Meiser-Stedman 2002). For example, carer-reported externalising, following maltreatment (Leeson Ehlers and Clark’s cognitive model of PTSD (Ehlers and and Nixon 2011). This is consistent with the centrality of Clark 2000) identifies three core posttrauma processes that maladaptive, dysfunctional or distorted cognitions about the maintain PTSD, particularly by contributing to a sense of cur- self, others, and/or world, to models of child depression (e.g., rent threat. First, the individual makes strong negative ap- Birmaher et al. 1996), anxiety (e.g., Ellis and Hudson 2010; praisals of the event or their own responses (e.g., BI’ll never Rapee and Heimberg 1997), and conduct problems (e.g., get over what happened^). Second, the trauma memory is Dodge and Pettit 2003). The role of posttrauma coping and proposed to be encoded in such a way that it is particularly memory quality in relation to broader psychopathology has prone to retrieval via direct triggering through matching cues been less well established. However, rumination has been as- and has ‘flashback’ like properties (e.g., due to a high level of sociated with non-trauma specific depression in children (e.g., sensory content, and limited contextual or semantic see Hitchcock et al. 2014), while avoidant coping is central to encoding). Third, engagement in maladaptive coping strate- the maintenance of a broad range of child internalising diffi- gies is hypothesised to maintain maladaptive appraisal and culties (e.g., Murray et al. 2009). Equally, researchers have trauma memory properties (e.g., due to cognitive avoidance), highlighted the presence of negative intrusive memories in and directly exacerbate symptoms (e.g., due to counterproduc- depression and in anxiety disorders (e.g., Hackmann and tive attempts to suppress trauma memories). Holmes 2004; Patel et al. 2007) The empirical literature, including a number of meta- The primary aim of the current study was to investigate the analytic reviews, provides strong support for the central role potential role of key trauma-related cognitive processes in of posttrauma cognitive processes in the maintenance of trau- relation to a range of psychological outcomes in children, in matic stress. For example, meta-analytic reviews of the child a longitudinal study of an acute-trauma exposed sample and adolescent trauma literature have concluded that the use (Hiller et al. 2017). We examined whether: a) trauma related of thought suppression (a maladaptive cognitive coping strat- appraisals, memory qualities, and maladaptive coping strate- egy) is moderately to strongly positively associated with gies showed cross-sectional associations with anxiety, depres- PTSS severity (Trickey et al. 2012), and that there is a strong sion and externalising symptoms, measured 1-month association between maladaptive appraisals and PTSS posttrauma, as well as with symptoms of PTSD; and b) wheth- (Mitchell et al. 2017; Trickey et al. 2012). Theroleof er changes in cognitive processes from 1 to 7 months were trauma-related memory quality is more well-established in associated with internalising and externalising severity across the adult PTSD field (Brewin 2014), although there is some this period. Given that there is overlap between symptoms of evidence for a significant association between sensory-laden PTSD and those of depression and anxiety problems, we also trauma memories and higher PTSS among children following examined whether posttrauma cognitive mechanisms made a trauma (Salmond et al. 2011;Meiser-Stedman et al. 2007a; unique contribution to the prediction of anxiety, depression, J Abnorm Child Psychol (2019) 47:683–693 685 and/or externalising symptoms once levels of PTSS were con- was assigned using the Manchester Triage System (nurse rat- trolled for. Although a particularly stringent analysis, this ing of urgency of care), ranging from 1 = immediate care allowed for stronger inferences in relation to whether required to 4 = standard (i.e., less urgent). Key study indices posttrauma psychological processes are relevant intervention were then obtained through child and parental questionnaires, targets even in the absence of significant PTSS. completed at both T1 and T2. Specifically, we utilised child self-report measures of cognitive processes, PTSS, anxiety and depressive symptoms. By contrast, we used parental re- Method port as our main measure of child externalising, as is recom- mended in the literature (Stanger and Lewis 1993). Participants Maladaptive Trauma Appraisals Children completed the Participants were 132 children aged 6–13 years old, and their Children’s Posttraumatic Cognitions Inventory (CPTCI), a 25- caregivers, recruited following the child’s involvement in a item self-report measure suitable for children as young as 6 years trauma and subsequent attendance at the emergency depart- old (Meiser-Stedman et al. 2009). The CPTCI measures ap- ment (ED). Participants were recruited between April 2014 praisals relating to: (i) permanent and disturbing change (e.g., and January 2016, from four EDs in the UK, as previously BMy reactions since the event mean I have changed for the reported (Hiller et al. 2017). Exclusion criteria were: signifi- worse^)and(ii) beingafragilepersonina scaryworld(e.g., cant learning difficulty or neurodevelopmental disorder; or- BAnyone could hurt me^). Young people rate their agreement ganic traumatic brain injury; suspicion that the injury was with each statement from 1 = don’t agree at all to 4 = agree a lot, caused by the young person themselves or their carer. with the total score providing an overall index of maladaptive Thirty-nine percent of all potentially eligible participants were appraisals (ranging from 25 to 100; α = 0.93). recruited to the final study (main reason for non-participation was that families could not be contacted). We found no evi- Posttrauma Coping Children completed the Child Posttrauma dence that study participants differed from the eligible popu- Coping Questionnaire (CPCQ), an 11-item self-report scale lation. Complete recruitment details are presented in Hiller created for this study based on cognitive coping items used et al. 2017, while the recruitment flow-chart is available in in previous research (Ehlers and Clark 2000;Ehlers supplementary materials (S1). Index traumas were: motor ve- et al. 2003; Stallard 2003). The measure includes 5 items hicle accident (n = 68, 52%), fall from a height (n = 25, 19%), on rumination (e.g., BIcan’t stop thinking if only the event significant bicycle accident (n = 9, 7%), acute medical episode hadn’t happened to me^) and 6 items on thought suppression (n = 10, 8%; e.g., acute anaphylaxis), sport injury (n =6, 5%), (e.g., BI’ve tried to keep any memories of what happened out and assault (n = 3, 2%) or other event (n = 17, 13%, e.g., house of my head^), each rated on a four-point scale from 0 = not at fire, dog attack, near drowning, sports injury). all or only one time to 3 = a lot of the time, and summed to yield a total score (between 0 and 30). The measure shows Procedure good internal consistency (α = 0.89) and validity against a measure of child PTSS symptoms (Hiller et al. 2017). ED staff initially approached families and obtained permission for the research team to make contact in order to confirm Trauma Memory Quality Children completed an adapted ver- eligibility and recruit them to the study. Informed consent sion of the Trauma Memory Quality Questionnaire was provided by the caregiver, while informed assent was (TMQQ; Meiser-Stedman et al. 2007a). This 18-item mea- provided by the young person. Data presented in the current sure covers the original scale’s 11-items on the sensory paper are from home assessments completed independently quality of the trauma-memory and sense of Bnowness^ by parents and children at 2–6 weeks post-ED attendance (e.g., BMy memories of the event are mostly pictures of (T1) and 6-months later (T2; i.e., 7-months posttrauma). images^; BWhen I think about the frightening event I can Dyads received GBP20 at each time point, as a thank you sometimes smell things that I smelt when the frightening for their time. An additional 3-month postal questionnaire event happened^), and an additional 7-items on assessment also took place, as reported in Hiller et al. 2017. disorganised memories (e.g., BI get mixed up about what Ninety-six percent of the sample were retained at the 6-month order things happened in during the frightening event^), assessment. adapted from an equivalent adult scale (e.g., Halligan et al. 2003). Items are rated on a scale from 1 = disagree a Measures lot to 4 = agree a lot and summedtoyieldatotal score (range, 18–72). Internal consistency of the 18-item scale Demographic and trauma-related information were obtained was α = 0.86, compared to α = 0.80 for the original 11- from ED notes and parent interview. Objective trauma severity item scale. 686 J Abnorm Child Psychol (2019) 47:683–693 PTSS Children completed the child self-report version of the controlling for PTSS. We used preliminary correlational anal- PTSD Reaction Index (PTSD-RI), which has established reli- yses to establish which cognitive variables showed significant ability and validity for children as young as 6 years old univariate associations with mental health outcomes (p <.05), (Steinberg et al. 2004). The PTSD-RI assesses 17 DSM-IV- for inclusion in subsequent regression analyses. To index T1- TR PTSD symptoms. Responses to each item are rated on a 5- T2 change in cognitions we generated residual change scores point Likert scale, ranging from 0 = none of the time to 4 = via linear regressions. Next, we ran separate linear regressions most of the time, and yielding a total symptom score as the to explore which of the three cognitive processes (measured at main outcome (ranging from 0 to 68; α =0.89). T1) most strongly predicted 1-month symptom severity (T1 PTSS, anxiety, depression, parent-reported internalising and Anxiety and Depression Children completed the 25-item short externalising), when they were examined simultaneously (i.e., version of the Revised Child Anxiety and Depression Scale entered into a single step in the regression). For non-PTSS (RCADS; Chorpitaetal. 2000; Ebesutani et al. 2012), a wide- outcomes, additional models tested for predictive effects after ly used, self-report measure, validated with children aged controlling co-occurring PTSS. Next, to explore the longitu- 7 years and over (Ebesutani et al. 2012). Items are rated on a dinal association between cognitions and symptoms, we 4-point scale ranging from 0 = never to 3 = always,measuring looked at whether change in cognitive processes was associ- 15 symptoms of anxiety (e.g., BI worry about being away from ated with change in symptoms, by running linear regression my parents^; total score range 0–45; α =0.89) and 10 depres- models with T2 symptoms as the dependent variable, whilst sive symptoms (e.g., BIfeel worthless^; total score range 0– controlling for T1 symptoms in the model. In particularly 30; α =0.86). conservative analyses, these regressions for non-PTSS out- comes were re-run also controlling for change in PTSS (resid- Parent-Report on Child Internalising and Externalising Parents ual change scores), to explore whether change in the cognitive completed the Strengths and Difficulties Questionnaire (SDQ) processes would continue to predict depression/anxiety/ parent-report version, a widely used measure of children’s internalising/externalising symptoms change, even above emotional and behavioural difficulties, suitable for reporting what is explained by PTSS. Age, sex and triage (a marker of on children as young as 4 years of age (Goodman 1997). objective trauma severity) were explored as potential Parents rate items from 0 = not true to 2 = certainly true.We covariates. utilised the 10-item externalising subscale which comprises Overall there was little missing data (maximum missing items covering hyperactivity and conduct problems (e.g., data was less than 10%). The only exception to this was the BOften has temper tantrums or hot tempers^, BEasily T1 SDQ, which was added after the study begun and was distracted^; total score range 0–20; α = 0.76); and the 10- missing for 20% of the sample. In all cases data were missing completely at random. As a sensitivity analysis, to account for item internalising scale, which assesses emotional and peer problems (e.g., BMany worries, often seems worried^; missing data, key analyses were run using multiple imputation BPicked on by other children^; total score range 0–20; α = with 50 iterations and predictive mean matching. The pattern 0.80). The SDQ was added to the T1 assessment during the of results was the same and completer-only data are presented course of the study; consequently, T1 data is only available for here. 106 participants. Data Analytic Plan Results Data were analysed using IBM SPSS Statistics for Descriptive Information Windows v22 (2013, IBM Corp., Armonk, NY). All mental- health outcome measure scores were positively skewed, so a Sample characteristics are presented in Table 1. The sample square-root transformation was applied. As most 6-month da- comprised 132 children (62% boys), aged 6–13 years old, and ta remained significantly skewed, associations were checked their participating parent (90% mothers). Child age, sex, and against non-parametric tests (Spearman’s rho), with any dis- triage category were assessed as potential covariates. The crepancies noted in Table 2. Our primary aims were to explore child’s age was negatively associated with PTSS at both time (1) whether the three cognitive processes (appraisals, memory, points (r > −0.18, p > 0.04). Sex was significantly associated coping) would be associated with acute (1-month, T1) symp- with initial anxiety (r = 0.21, p = 0.04) and depression (r = tom severity, and (2) whether change in these processes from 1 0.20, p = 0.049) symptom severity, with higher symptom to 7 months posttrauma (T1 – T2) would be associated with scores associated with being female. Triage ratings were not change in symptoms over this period. In all cases, we also significantly associated with any measures of mental health examined whether cognitive processes would continue to be (all ps > 0.13). Age and sex were controlled in all subsequent significantly associated with symptom outcomes, even after analyses. J Abnorm Child Psychol (2019) 47:683–693 687 Table 1 Descriptive information (appraisals β =0.43, p <0.001; memory β =0.15, p =0.034; coping β =0.17, p =0.018). Demographic characteristics Statistic (N =132) Parent characteristics Anxiety Symptoms Age in years (M[SD]) 39.7 (7.0) Proportion mothers 119 (90%) Each cognitive process was significantly positively correlated Proportion married or cohabiting 97 (74%) with child reported anxiety at T1, and change in each process was associated with equivalent change in anxiety scores from Education status: School until 16 years or younger 36 (27%) Further education 50 (38%) T1 to T2 (see Tables 2 and 3). In a regression model control- ling for age and sex, combined T1 cognitive processes ex- Higher education 46 (35%) plained 63% of variance in T1 anxiety scores, with each being Child characteristics a unique predictor (see Table 4). When additionally control- Age in years, M(SD) 9.87 (1.8) ling for T1 PTSS, the combined cognitive processes still ex- Male 82 (62.1%) plained 12% of variance in T1 anxiety, and maladaptive ap- Ethnicity – Caucasian 121 (91.7%) praisals and coping remained unique predictors (see Table 4). Triage category We next tested whether change in cognitive processes sig- 1 (immediate attention required) 61 (46%) nificantly predicted change in anxiety from T1 to T2, by con- 2 (very urgent) 29 (22%) trolling for T1 anxiety scores (and age and sex) in a regression 3 (urgent) 26 (20%) model with T2 anxiety as the dependent variable. In combina- 4 (less urgent) 16 (12%) tion, cognitive change scores explained 28% of variance in this Days in hospital (Min – Max, M [SD]) 0–28, 2.64 (4.83) a model, with change in maladaptive appraisals and cognitive Days of school missed (Min – Max, M [SD]) 0–28, 5.52 (5.98) coping, but not trauma-memory, uniquely predicting change Proportion requiring ambulance/helicopter 90 (70%) in anxiety symptoms across the 6-months (see Table 5). Proportion with head injury 33 (25%) When additionally controlling for change in PTSS, combined Days of school missed represents days missed prior to their first cognitive processes remained a significant predictor of anxiety, assessment explaining 3% of variance, although no individual cognitive processes uniquely predicted anxiety change (Table 5). Paired samples t-tests indicated that children’s self-reported symptoms significantly reduced from 1-month to 6-months Depressive Symptoms across all measures (M and SDs presented in Table 2; PTSS, p < 0.001; Anxiety, p< 0.001; Depression, p < 0.001). By con- Each cognitive process showed significant positive associa- trast, there was no significant change in parent-report of child tions with T1 depression and with change in depression from internalising (p = 0.75) or externalising symptoms (p = 0.47). T1 to T2 (see Tables 2 and 3). In a regression analysis, com- bined T1 cognitive processes explained 59% of variance in T1 depression, F(3,114) = 62.64, p < 0.001, with all three pro- Posttraumatic Stress Symptoms cesses being significant independent predictors (see Table 4). When additionally controlling for T1 PTSS, cognitive pro- Significant bivariate correlations were present between all cesses explained 5% of the variance in depression scores, cognitive processes (i.e., appraisals, trauma memory, cogni- and maladaptive appraisals remained uniquely associated with tive coping) and PTSS at T1, and change in each process was depression severity (Table 4). significantly associated with change in PTSS scores from T1 Next, after controlling for initial depression severity (and to T2 (see correlation matrices in Tables 2 and 3). In a regres- age and sex), change in cognitions from T1 to T2 significantly sion model controlling for age and sex, the combined T1 cog- predicted T2 depression scores, explaining 28% of variance. nitive processes explained 72% of variance in T1 PTSS, Here change in maladaptive appraisals and cognitive coping F(3,114) = 115.98, p < 0.001, with appraisals (β = 0.30, both uniquely predicted depression severity. However, after p < 0.001), memory (β =0.33, p < 0.001) and coping (β = additionally controlling for change in PTSS scores, change 0.41, p < 0.001) each being unique predictors in the model. in cognitive processes no longer significantly predicted We next tested whether change in cognitive processes predict- changes in depression (see Table 5). ed change in PTSS, by examining prediction of T2 PTSS by cognitive change scores, in a regression model controlling for Parent-Reported Child Internalising Symptoms T1 PTSS (as well as age, sex). Combined cognitive processes explained 41% of variance, F(3,102) = 43.11, p <0.001, and Each cognitive process was correlated with parent-reported each remained a unique predictor in this longitudinal model child internalising scores at T1, but only maladaptive 688 J Abnorm Child Psychol (2019) 47:683–693 Table 2 Bivariate correlation matrix for associations between study variables 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 1-mo Processes 1. Appraisals 2. Coping 0.61** 3. Memory 0.59** 0.63** 1-mo Outcomes 4. PTSS 0.68** 0.79** 0.75** 5. Anxiety 0.74** 0.68** 0.61** 0.74** 6. Depression 0.70** 0.69** 0.58** 0.79** 0.84** a + 7. Internalising 0.37** 0.23* 0.22* 0.25** 0.18 0.24* 8. Externalising 0.29** 0.09 0.03 0.21 0.24* 0.29** 0.51** 6-mo Outcomes 9. PTSS 0.59** 0.54** 0.39** 0.62** 0.68** 0.68** 0.21* 0.30** 10. Anxiety 0.62** 0.43** 0.35** 0.54** 0.67** 0.64** 0.12 0.29** 0.75** 11. Depression 0.61** 0.45** 0.30** 0.54** 0.67** 0.71** 0.25* 0.34** 0.79** 0.84** a + 12. Internalising 0.28** 0.13 0.17 0.29** 0.36** 0.41** 0.53** 0.55** 0.49** 0.39** 0.55** 13. Externalising 0.23* −0.003 0.06 0.13 0.26* 0.21 0.33** 0.61** 0.29** 0.27* 0.39** 0.67** M (SD) 40.20 14.52 37.51 18.84 9.65 6.30 3.84 6.01 12.86 6.67 4.08 3.72 6.27 (13.47) (8.86) (10.31) (13.54) (8.40) (6.67) (3.71) (3.89) (11.88) (7.18) (4.82) (4.05) (4.79) Using Spearman’s rho, 1-mo parent-report internalising was no longer significantly correlated with coping (r =0.20, p =0.06) but 6-mo parent- reported internalising was significantly correlated with coping (r =0.26, p = 0.02). Coping was retained in the regression models. There were no other discrepancies between parametric (Bivariate) and non-parametric (Spearman’sRho) tests p < 0.10, *p< 0.05, **p < 0.001. PTSS = child-reported posttraumatic stress symptom severity. ‘Internalising’ and ‘Externalising’ based on parent- report on the SDQ, all other questionnaires were child self-report. Possible ranges of scores for each measure are presented in the Method appraisals and coping (not memory) were significantly asso- As only change in appraisals and coping from T1 to ciated with change in symptoms from T1 to T2 (see Table 3). T2 were significantly associated with change in At T1, in combination, the acute cognitive processes ex- internalising severity (Table 3), only these processes plained 15% of variance in parent-reported internalising, with were examined further in regression models. In a model maladaptive appraisals being a significant unique predictor controlling for initial internalising scores, cognitive (see Table 4). After additionally controlling for T1 PTSS, change scores in combination explained a significant cognitive processes no longer explained significant variance 10% of variance in 6-month internalising problems. in T1 parent-report internalising, although maladaptive ap- Here, change in the child’s cognitive coping was praisals remained a unique significant predictor (Table 4). uniquely associated with change in internalising Table 3 Bivariate correlations between residual change scores 1. 2. 3. 4. 5. 6. 7. 1. Appraisals 2. Coping 0.59** 3. Memory 0.55** 0.50** 4. PTSS 0.77** 0.56** 0.56** 5. Anxiety 0.63** 0.51** 0.41** 0.67** 6. Depression 0.61** 0.47** 0.36** 0.67** 0.72** 7. Internalising 0.25* 0.33** 0.16 0.28** 0.31** 0.40** 8. Externalising 0.22* 0.19 0.08 0.12 0.13 0.27** 0.52** Using Spearman’s Rho, change in appraisals was not significantly associated with change in externalising (r= 0.12, p = 0.29). There were no other discrepancies between parametric (Bivariate) and non-parametric (Spearman’sRho) tests *p <0.05, ** p < 0.01. All scores are 1–6 month residual change scores. Internalising and Externalising based on parent report, all other scores based on child self-report J Abnorm Child Psychol (2019) 47:683–693 689 Table 4 Results of linear regressions for T1 cognitive predictors of T1 Table 5 Results of linear regressions analyses of longitudinal data symptoms, after controlling for age and sex (Model 1) and T1 PTSS examining change in cognitive processes as a predictor of symptom (Model 2) change (Model 1), including when controlling for change in PTSS (Model 2) Model 1 Model 2 Model 1 Model 2 Child Report 2 2 Child Report Anxiety R =0.63, R =0.12, 2 2 F(3,114) = 77.6** F(3,113) = 14.7** Anxiety R =0.28, R =0.03, F(3,102) = 22.4** F(3,101) = 3.0* Appraisals 0.51** 0.45** Appraisals 0.38** 0.19 Memory 0.16* 0.10 Memory 0.02 −0.03 Coping 0.28** 0.21** 2 2 Coping 0.20* 0.15 Depression R =0.59, R =0.05, 2 2 F(3,114) = 62.6*** F(3,113) = 6.04*** Depression R =0.28, R =0.03, F(3,102) = 23.3*** F(3,101) = 2.4 Appraisals 0.44** 0.29** Appraisals 0.41** 0.17 Memory 0.19* 0.06 Memory 0.003 −0.06 Coping 0.30** 0.15 Coping 0.19* 0.12 Parent Report 2 2 + Parent Report Internalising R =0.15, R =0.06, F(3,91) = 2.2 2 2 F(3,92) = 5.4** Internalising R =0.10, F(2,81) = 6.6** R =0.03, F(2,80) = 2.2 Appraisals 0.36** 0.34* Appraisals 0.03 −0.16 Memory 0.06 0.03 Coping 0.30* 0.24* 2 + 2 Coping −.01 −0.05 Externalising R =0.02,F(1,82) = 3.4 R =0.01,F(1,81) = 2.0 2 2 Externalising R =0.07, R =0.07,F(1,93) = 4.2 Appraisals 0.16 0.19 F(1,94) = 7.5** Appraisals 0.28** 0.28* R and F statistics relate to the change in model fit following the inclusion of cognitive variables. For the change models, we examined whether T1 R and F statistics relate to the change in model fit following the inclusion to T2 residualized change scores for cognitive processes predicted T2 of cognitive variables, controlling for age and sex of the child (Model 1) symptom severity, controlling for age, sex and T1 symptoms (Model 1), and additionally for PTSS (Model 2) and then once also controlling for residualized change in PTSS (Model 2) ** p ≤ 0.01, *p<0.05, p < 0.10 ** p ≤ 0.01, *p<0.05, p < 0.10 severity, but appraisal change was not (Table 5). After controlling for change in PTSS, combined cognitive Discussion processes no longer explained significant variance, al- though change in cognitive coping remained a unique We examined trauma-related psychological processes (ap- significant predictor of change in parent-report praisals, coping, memory) as potential predictors of children’s internalising (Table 5). broader psychological outcomes following exposure to acute trauma. We replicated findings that all three processes explain unique variance in acute PTSS and reductions in PTSS over Parent-Reported Child Externalising Symptoms time. We also found robust evidence that children’s posttrauma cognitive processes were cross-sectionally and Only maladaptive appraisals showed significant bivariate longitudinally associated with non-PTSD internalising prob- lems, based on both self- and parent-report. These associations associations with externalising symptoms at either T1 or with change at T2, and was considered in regression with wider child internalising symptoms were particularly re- models (see Tables 2 and 3). After controlling for age liable for child-reported maladaptive appraisals and maladap- and sex, initial maladaptive appraisal scores were signif- tive cognitive coping strategies (i.e., thought suppression and icantly associated with parent-reported externalising rumination); and were partially retained even after levels of symptoms at T1, explaining 7% of variance. This effect PTSS were controlled for. There was less evidence for the role remained significant after controlling for concurrent 1- of trauma related cognitive processes in predicting externalising difficulties. month PTSS, explaining 4% of variance in parent- reported child externalising symptoms (see Table 4). Our findings add to the limited literature showing trauma- related cognitive processes are important for children’s However, after controlling for initial externalising symp- toms (and age and sex), change in maladaptive ap- broader mental health following trauma. In particular, it builds on cross-sectional studies that have demonstrated associations praisals from T1 to T2 was not significantly predictive of T2 externalising (Table 5). between trauma-related negative appraisals and broader 690 J Abnorm Child Psychol (2019) 47:683–693 psychopathology (e.g., Leeson and Nixon 2011; Liu and Chen was no evidence that memory quality predicted wider 2015). The current study found that maladaptive appraisals, internalising problems once PTSS was controlled for. Caution coping and trauma memory quality at 1-month posttrauma is warranted when interpreting these null findings, due to the were each moderately to strongly correlated with children’s strong overlap between all three processes and sets of symp- self-reported depression and anxiety symptoms, and showed toms. Nonetheless, in this case we cannot rule out the possibil- slightly smaller associations with parent-reported child ity that associations with anxiety and depression are accounted internalising symptoms. Each cognitive process explained for by the co-occurrence of PTSD symptomatology. Whereas unique variance in child reported non-PTSD internalising negative appraisals and maladaptive coping strategies are wide- symptoms at 1-month posttrauma, whereas negative ap- ly implicated in models of anxiety (e.g., thought suppression) praisals were the only independent predictor of parent- and depression (rumination), trauma-memory quality may be a reported internalising at this timepoint. Longitudinal analyses unique driver of traumatic stress. This is consistent with theo- examining change in cognitive processes over time found a retical perspectives that highlight traumatic memories as a key similar pattern of results. Together, changes in maladaptive defining feature of PTSD versus other emotional disorders, cognitive processes explained a relatively large 28% of vari- both phenomenologically and in terms of underlying neural ance in change in anxiety and depression across a 6-month processes (e.g., Elzinga and Bremner 2002). follow-up, and a smaller 10% of variance in change in parent- Overall, our evidence highlights the potential utility of con- reported internalising problems. Changes in appraisals and sidering trauma-related maladaptive coping strategies and ap- coping were identified as uniquely associated with symptom praisals as potential treatment targets in depressed or anxious change when cognitive processes were examined simulta- youth who present following trauma exposure. That the same neously as predictors. processes that maintain PTSS also contribute to non-PTSD Although not the main focus of the current study, we also internalising may also explain why many CBT trials targeting found robust longitudinal evidence that appraisals, memory PTSD see concurrent reductions in internalising comorbidities and maladaptive coping strategies each explain unique variance (e.g., Deblinger et al. 2011; Goldbeck et al. 2016; Smith et al. in children’s PTSS, consistent with cognitive models (e.g., 2007). Our findings suggest that this change in broader psycho- Ehlers and Clark 2000). In particularly stringent analyses, we pathology likely particularly results from the targeting of chil- examined whether cognitive processes could predict broader dren’s maladaptive appraisals and coping strategies. While our internalising symptoms when controlling for co-occurring findings suggest that a robust focus on appraisals and maladap- PTSS. This substantially reduced the variance explained by tive coping may be particularly clinically useful when young cognitive factors, which is unsurprising given the strong corre- people present with internalising problems linked to trauma, lation and symptom overlap between PTSS, anxiety and de- replication is required before drawing clinical conclusions. pression. Nevertheless, even after controlling for concurrent We found substantially less evidence that posttrauma cog- PTSS we still found that maladaptive appraisals (e.g., Bthe nitive processes contribute to child externalising symptoms. world is unsafe^, BI’ll never get over what happened^)and Our findings build on Liu and Chen’s(2015) study with a cognitive coping (i.e., thought suppression, rumination), community-sample of adolescents, replicating a cross- remained cross-sectionally associated with children’s non- sectional association between appraisals and externalising, PTSD internalising at 1-month posttrauma. Moreover, in lon- but additionally suggest that changes in trauma-related ap- gitudinal analyses, reductions in maladaptive cognitive coping praisals do not drive longer-term adjustment in terms of over time continued uniquely to predict change in parent- externalising behaviour. We found no evidence that trauma reported child internalising problems, while the combined cog- memory quality or maladaptive cognitive coping relate to chil- nitive processes significantly predicted reductions in child re- dren’s posttrauma externalising symptoms. The suggestion ported anxiety. This indicates that associations are unlikely to that these psychological processes may be less relevant to be secondary to PTSD symptoms, but rather that cognitive the maintenance of child externalising is consistent with some processes potentially make a direct contribution to children’s findings from treatment studies. For example, Deblinger et al. wider internalising problems following trauma. (2011) found that externalising was more effectively targeted It is notable that whereas negative appraisals and maladap- in a treatment that had a larger focus on parent-training, com- tive coping strategies were relatively consistently associated pared to a treatment that placed more focus on children’s with children’s anxiety and depression, there was less robust trauma-related memories and appraisals. Thus, characteristics evidence for the role of trauma-related memory qualities in of children’s posttrauma social environment may be more rel- relation to longer-term broader psychopathology. Thus, al- evant to managing posttrauma externalising behaviours though trauma memory quality showed a clear pattern of bivar- (Deblinger et al. 1996; Silverman et al. 2008). iate associations with child internalising symptoms, it did not Findings should be interpreted in the light of limitations. emerge reliably as a unique predictor of distress over and above First, as this is an observational longitudinal study, causation other processes, particularly in longitudinal models; and there cannot be determined. Second, we did not include a child- J Abnorm Child Psychol (2019) 47:683–693 691 and RMS is an NIHR Career Development Fellow (CDF-2015-08-073). report measure of externalising symptoms in our study, and we The views expressed are those of the authors and not necessarily those of were unable to compare exact measures of internalising (i.e., the NHS, the NIHR or the Department of Health. anxiety and depression) across child- versus parent-report. The approach we took to symptom measurement is consistent with Funding This research was funded by an Economic and Social Research best practice: parent or teacher reports are considered the gold Council (ESRC) grant awarded to SLH (ES/K006290/1). standard for measuring children’s externalising (Stanger and Lewis 1993), whereas child-report is recommended for mea- Compliance with Ethical Standards surement of PTSS (e.g., Kassam-Adams et al. 2006;Meiser- Conflicts of Interest None. Stedman et al. 2007b). Our observation that posttrauma child internalising problems reduced from 1-month to 6-months ac- Ethical Approval Ethical approval was obtained from the University of cording to child-report but not parent-report may suggest that Bath Research Ethics Committee and the NHS Research Ethics parents are reporting on children’s pre-established psychologi- Committee South Central - Oxford A (Ref 137454). cal profiles, rather than detecting changes in their posttrauma mental health. This is consistent with the view that parents may Data Access Halligan, Sarah and Hiller, Rachel (2017). The role of trauma specific behaviours and parenting style in facilitating child psy- be relatively poor at detecting children’s posttrauma symptoms. chological adjustment. [Data Collection]. Colchester, Essex: UK Data Parental perceptions of their child’s distress may also be influ- Archive https://doi.org/10.5255/UKDA-SN-852668 enced by their own posttrauma distress. Nonetheless, single- Open Access This article is distributed under the terms of the Creative informant bias may substantially inflate associations in the con- Commons Attribution 4.0 International License (http:// text of psychopathology, and associations between child- creativecommons.org/licenses/by/4.0/), which permits unrestricted use, reported symptoms and cognitions should be considered with distribution, and reproduction in any medium, provided you give this in mind. Null findings should also be considered in the appropriate credit to the original author(s) and the source, provide a link light of the relatively modest overall sample size. The sample to the Creative Commons license, and indicate if changes were made. size also meant we were not appropriately powered to run more sophisticated analyses, such as structural equation models, which may be better able to account for covariance between References the different symptom outcomes. Finally, it is notable that, as is typical of low-risk acute-incident emergency department sam- Alisic, E., Zalta, A. K., Van Wesel, F., Larsen, S. E., Hafstad, G. S., ples, overall internalising and externalising symptoms in our Hassanpour, K., & Smid, G. E. (2014). Rates of posttraumatic stress sample were relatively low. For the majority of participants the disorder in trauma-exposed children and adolescents: Meta-analysis. The British Journal of Psychiatry, 204,335–340. index trauma was an acute accidental injury, and while families Birmaher, B., Ryan, N. D., Williamson, D. E., Brent, D. A., Kaufman, J., were from a range of socio-economic backgrounds, the sample Dahl, R. E., et al. (1996). Childhood and adolescent depression: A was mostly culturally- and ethnically-homogenous, and cap- review of the past 10 years. Part I. Journal of the American tured children up to 13 years old. Findings cannot be general- Academy of Child & Adolescent Psychiatry, 35(11), 1427–1439. ised to more complex traumas or high-risk groups, or to ado- Brewin, C. R. (2014). Episodic memory, perceptual memory, and their interaction: Foundations for a theory of posttraumatic stress disor- lescents or other cultural contexts. der. Psychological Bulletin, 140(1), 69. In conclusion, cognitive processes that are known to be Brewin, C. R., Dalgleish, T., & Joseph, S. (1996). A dual representation important in the maintenance of PTSD are also likely to be theory of posttraumatic stress disorder. Psychological Review, relevant for children’s broader posttrauma internalising diffi- 103(4), 670–686. culties. Our findings particularly highlight the role of mal- Cénat, J. M., & Derivois, D. (2015). Long-term outcomes among child and adolescent survivors of the 2010 Haitian earthquake. adaptive coping and maladaptive appraisals, in predicting Depression and Anxiety, 32(1), 57–63. acute distress and reductions in non-PTSD internalising prob- Chorpita, B. F., Yim, L., Moffitt, C., Umemoto, L. A., & Francis, S. E. lems, and suggest that it may be beneficial to take a trauma- (2000). Assessment of symptoms of DSM-IV anxiety and depres- informed approach to addressing posttrauma internalising sion in children: A revised child anxiety and depression scale. problems by targeting these processes. What remains to be Behaviour Research and Therapy, 38(8), 835–855. Cohen, J. A., Mannarino, A. P., Berliner, L., & Deblinger, E. (2000). seen is whether trauma informed treatment frameworks may Trauma-focused cognitive behavioral therapy for children and ado- effectively target trauma-related anxiety and depression in the lescents: An empirical update. Journal of Interpersonal Violence, absence of clinical-level PTSD. 15(11), 1202–1223. Deblinger, E., Lippman, J., & Steer, R. (1996). Sexually abused children Acknowledgements We thank families for their generosity in participat- suffering posttraumatic stress symptoms: Initial treatment outcome ing; clinical teams at the Bristol Royal Hospital for Children, Royal findings. Child Maltreatment, 1,310–321. United Hospital Bath, Great Western Hospital, and the Gloucestershire Deblinger, E., Mannarino, A. P., Cohen, J. A., Runyon, M. K., & Steer, R. Royal Hospital for supporting our research; and Hope Christie, Claire A. (2011). Trauma-focused cognitive behavioral therapy for chil- Dixon and Victoria Williamson for their assistance with this research. dren: Impact of the trauma narrative and treatment length. CC is funded by an NIHR Research Professorship (RP_2014-04-018) Depression and Anxiety, 28(1), 67–75. 692 J Abnorm Child Psychol (2019) 47:683–693 Dodge, K. A., & Pettit, G. S. (2003). A biopsychosocial model of the psychopathology in Taiwanese children and adolescents. Journal of Abnormal Child Psychology, 43(8), 1475–1484. development of chronic conduct problems in adolescence. Developmental Psychology, 39(2), 349. Mannarino, A. P., & Cohen, J. A. (1996). Abuse-related attributions and Ebesutani, C., Reise, S. P., Chorpita, B. F., Ale, C., Regan, J., Young, J., perceptions, general attributions, and locus of control in sexually Higa-McMillan, C., & Weisz, J. R. (2012). The revised child anxiety abused girls. Journal of Interpersonal Violence, 11(2), 162–180. and depression scale-short version: Scale reduction via exploratory McKinnon, A., Brewer, N., Cameron, K., & Nixon, R. D. (2017). The bifactor modeling of the broad anxiety factor. Psychological relationship between processing style, trauma memory processes, Assessment, 24(4), 833. and the development of posttraumatic stress symptoms in children and adolescents. Journal of Behavior Therapy and Experimental Ehlers, A., & Clark, D. M. (2000). A cognitive model of posttraumatic Psychiatry, 57,135–142. stress disorder. Behaviour Research and Therapy, 38,319–345. Meiser-Stedman, R. (2002). Towards a cognitive–behavioral model of Ehlers, A., Mayou, R. A., & Bryant, B. (2003). Cognitive predictors of PTSD in children and adolescents. Clinical Child and Family posttraumatic stress disorder in children: Results of a prospective Psychology Review, 5(4), 217–232. longitudinal study. Behaviour Research and Therapy, 41(1), 1–10. Meiser-Stedman, R., Dalgleish, T., Smith, P., Yule, W., & Glucksman, E. Ellis, D. M., & Hudson, J. L. (2010). The metacognitive model of gen- (2007a). Diagnostic, demographic, memory quality, and cognitive eralized anxiety disorder in children and adolescents. Clinical Child variables associated with acute stress disorder in children and ado- and Family Psychology Review, 13(2), 151–163. lescents. Journal of Abnormal Psychology, 116(1), 65–79. Elzinga, B. M., & Bremner, J. D. (2002). Are the neural substrates of Meiser-Stedman, R., Smith, P., Glucksman, E., Yule, W., & Dalgleish, T. memory the final common pathway in posttraumatic stress disorder (2007b). Parent and child agreement for acute stress disorder, post- (PTSD)? Journal of Affective Disorders, 70(1), 1–17. traumatic stress disorder and other psychopathology in a prospective Foa, E. B., Steketee, G., & Rothbaum, B. O. (1989). Behavioral/cognitive study of children and adolescents exposed to single-event trauma. conceptualizations of post-traumatic stress disorder. Behavior Journal of Abnormal Child Psychology, 35(2), 191–201. Therapy, 20,155–176. Meiser-Stedman, R., Smith, P., Bryant, R., Salmon, K., Yule, W., Goldbeck, L., Muche, R., Sachser, C., Tutus, D., & Rosner, R. (2016). Dalgleish, T., & Nixon, R. D. (2009). Development and validation Effectiveness of trauma-focused cognitive behavioral therapy for of the child post-traumatic cognitions inventory (CPTCI). Journal of children and adolescents: A randomized controlled trial in eight Child Psychology and Psychiatry, 50(4), 432–440. German mental health clinics. Psychotherapy and Psychosomatics, Mitchell, R., Brennan, K., Curran, D., Hanna, D., & Dyer, K. F. (2017). A 85(3), 159–170. meta-analysis of the association between appraisals of trauma and Goodman, R. (1997). The strengths and difficulties questionnaire: A re- posttraumatic stress in children and adolescents. Journal of search note. Journal of Child Psychology and Psychiatry, 38,581–586. Traumatic Stress, 30(1), 88–93. Hackmann, A., & Holmes, E. (2004). Reflecting on imagery: A clinical Murray, L., Creswell, C., & Cooper, P. J. (2009). The development of perspective and overview of the special issue of memory on mental anxiety disorders in childhood: An integrative review. imagery and memory in psychopathology. Memory, 12(4), 389–402. Psychological Medicine, 39(9), 1413–1423. Halligan,S.L.,Michael,T.,Clark,D.M.,&Ehlers,A.(2003). Patel, T., Brewin, C. R., Wheatley, J., Wells, A., Fisher, P., & Myers, S. Posttraumatic stress disorder following assault: The role of cognitive (2007). Intrusive images and memories in major depression. processing, trauma memory, and appraisals. Journal of Consulting Behaviour Research and Therapy, 45(11), 2573–2580. and Clinical Psychology, 71(3), 419. Perrin, S., Smith, P., & Yule, W. (2000). Practitioner review: The assess- Hiller, R. M., Meiser-Stedman, R., Fearon, P., Lobo, S., McKinnon, A., ment and treatment of posttraumatic stress disorder in children and Fraser, A., & Halligan, S. L. (2016). Research review: Changes in adolescents. The Journal of Child Psychology and Psychiatry and the prevalence and symptom severity of child post-traumatic stress Allied Disciplines, 41(3), 277–289. disorder in the year following trauma–a meta-analytic study. Journal Pine, D. S., & Cohen, J. A. (2002). Trauma in children and adolescents: of Child Psychology and Psychiatry, 57,884–898. Risk and treatment of psychiatric sequelae. Biological Psychiatry, Hiller, R. M., Meiser-Stedman, R., Lobo, S., Creswell, C., Fearon, P., 51(7), 519–531. Ehlers, A., Murray, L., Halligan, S. L. (2017). The role of parenting Rapee, R. M., & Heimberg, R. G. (1997). A cognitive-behavioral model style and behaviours in facilitating child posttrauma psychological of anxiety in social phobia. Behaviour Research and Therapy, 35(8), adjustment. Journal of Child Psychology and Psychiatry. In Press. 741–756. Hitchcock, C., Nixon, R. D., & Weber, N. (2014). A review of Salmond, C. H., Meiser-Stedman, R., Glucksman, E., Thompson, P., overgeneral memory in child psychopathology. British Journal of Dalgleish, T., & Smith, P. (2011). The nature of trauma memories Clinical Psychology, 53(2), 170–193. in acute stress disorder in children and adolescents. Journal of Child Kassam-Adams, N., Garcia-Espana, J. F., Miller, V. A., & Winston, F. Psychology and Psychiatry, 52(5), 560–570. (2006). Parent-child agreement regarding children_s acute stress: Scheeringa,M. S.,&Zeanah,C.H.(2008).Reconsiderationof The role of parent acute stress reactions. Journal of the American harm’s way: Onsets and comorbidity patterns of disorders in Academy of Child and Adolescent Psychiatry, 45(12), 1485–1493. preschool children and their caregivers following hurricane Le Brocque, R. M., Hendrikz, J., & Kenardy, J. A. (2009). The course of Katrina. Journal of Clinical Child and Adolescent posttraumatic stress in children: Examination of recovery trajecto- Psychology, 37,508–518. ries following traumatic injury. Journal of Pediatric Psychology, Scheeringa, M. S. (2015). Untangling psychiatric comorbidity in young 35(6), 637–645. children Who experienced single, repeated, or Hurricane Katrina Leeson, F. J., & Nixon, R. D. (2011). The role of children’s appraisals on traumatic events. Child Youth Care Forum, 44, 475. https://doi. adjustment following psychological maltreatment: A pilot study. org/10.1007/s10566-014-9293-7. Journal of Abnormal Child Psychology, 39(5), 759–771. Silverman, W. K., Ortiz, C. D., Viswesvaran, C., Burns, B. J., Kolko, D. Liu, S. T., & Chen, S. H. (2015). A community study on the relationship J., Putnam, F. W., & Amaya-Jackson, L. (2008). Evidence-based of posttraumatic cognitions to internalizing and externalizing psychosocial treatments for children and adolescents exposed to J Abnorm Child Psychol (2019) 47:683–693 693 traumatic events. JournalofClinicalChild &Adolescent Stanger, C., & Lewis, M. (1993). Agreement among parents, teachers, and children on internalizing and externalizing behavior problems. Psychology, 37(1), 156–183. Smith, P., Yule, W., Perrin, S., Tranah, T., Dalgleish, T., & Clark, D. M. Journal of Clinical Child Psychology, 22(1),107–116. (2007). Cognitive-behavioral therapy for PTSD in children and ad- Steinberg, A. M., Brymer, M. J., Decker, K. B., & Pynoos, R. S. (2004). olescents: A preliminary randomized controlled trial. Journalofthe The University of California at Los Angeles post-traumatic stress American Academy of Child & Adolescent Psychiatry, 46,1051– disorder reaction index. Current Psychiatry Reports, 6(2),96–100. Trickey, D., Siddaway, A. P., Meiser-Stedman, R., Serpell, L., & Field, A. Stallard, P. (2003). A retrospective analysis to explore the applicability of P. (2012). A meta-analysis of risk factors for posttraumatic stress the Ehlers And Clark (2000) cognitive model to explain PTSD in disorder in children and adolescents. Clinical Psychology Review, children. Behavioural and Cognitive Psychotherapy, 31(3),337– 32,122–138.

Journal

Journal of Abnormal Child Psychology – Springer Journals

Published: Sep 28, 2018

There are no references for this article.