Prospective Evaluation of Parent Distress Following Pediatric Burns and Identification of Risk Factors for Young Child and Parent Posttraumatic Stress Disorder

Abstract

Objective:

Early childhood is a high-risk time for exposure to potentially traumatic medical events. We have previously reported that 10% of young children continue to have posttraumatic stress disorder (PTSD) 6 months after burn injury. This study aimed to 1) document the prevalence and prospective change in parental psychological distress over 6 months following their child's burn injury and 2) identify risk factors for posttraumatic stress symptoms (PTSS) in young children and their parents.

Methods:

Participants were 120 parents of 1–6-year-old children with unintentional burn injuries. Data were collected within 2 weeks, 1 month, and 6 months of burn injury using developmentally sensitive diagnostic interviews and questionnaires. Results: Within the first month, ∼25% of parents had a probable PTSD diagnosis, and moderate to extremely severe levels of depression, anxiety, and stress. Distress levels decreased significantly over time; however, 5% of parents still had probable PTSD at 6 months. Hierarchical multiple regression and path analyses indicated that parent posttraumatic stress reactions contributed significantly to the development and maintenance of child PTSS. Other risk factors for child PTSS included premorbid emotional and behavioral difficulties and larger burn size. Risk factors identified for parent PTSS included prior trauma history, acute distress, greater number of child invasive procedures, guilt, and child PTSS.

Conclusions:

The findings from this study suggest that parents' responses to a traumatic event may play a particularly important role in a young child's psychological recovery. However, further research is needed to confirm the direction of the relationship between child and parent distress. This study identified variables that could be incorporated into screening tools or targeted by early intervention protocols to prevent the development of persistent child and parent PTSS following medical trauma.

Introduction

Infants, toddlers, and preschoolers typically fall within the highest risk category for exposure to sexual and physical abuse, unintentional injury, and witnessing domestic violence (Australian Institute of Health and Welfare 2009; United States Department of Health and Services 2009). Although the majority of young children who are exposed to trauma are likely to naturally recover over time, a significant proportion (6.5–50%) develop posttraumatic stress disorder (PTSD) and other comorbid conditions, and are at risk of following a chronic and debilitating trajectory (De Young et al. 2011b). Because of the unique developmental factors associated with early childhood (e.g., rapid rate of neurobiological development, limited emotion regulation skills, importance of a protective attachment relationship), the risk of adverse psychological and physiological consequences may be even more long lasting and widespread than for older children and adults (Perry et al. 1995; Carpenter and Stacks 2009).

Parents are often directly involved in their child's traumatic event and are also at risk of developing PTSD, anxiety, or depression (Scheeringa and Zeanah 2008; Bakker et al. 2013). Research has consistently documented a significant positive association between preschool posttraumatic stress symptoms (PTSS) and parent distress (Stoddard et al. 2006; Scheeringa and Zeanah 2008; Graf et al. 2011). The parent–child relationship is particularly important during early childhood (Lieberman 2004; Carpenter and Stacks 2009); therefore, there is a great need to understand the mechanisms influencing the relationship between child and parent distress as well as identifying what other variables influence child and parent outcomes following trauma.

Scheeringa and Zeanah (2001) have proposed a bidirectional model, termed “relational PTSD,” to describe the relationship between child and parent distress. This conceptual model suggests that trauma affects both the child and parent, and that each person's symptomatology exacerbates that of the other. Three suggested ways in which parents may influence their child's recovery are by becoming 1) withdrawn, unresponsive or unavailable, 2) overprotective or constricting, or 3) frightening, re-enacting, or endangering (Scheeringa and Zeanah 2001). However, longitudinal research is needed to test these hypotheses and to determine the direction of effects between child and parent symptomatology following trauma.

The empirical and theoretical trauma risk factor literature suggests that it also important to consider other pretrauma, trauma-related and posttrauma individual and environmental variables that may influence the development of PTSS in children and parents. Research is sparse with infants and preschoolers; however, some potential risk factors that have been identified for PTSS in this age group following injury include pretrauma externalizing behavior and witnessing a threat to their caregiver (Scheeringa et al. 2006), greater burn size, number of invasive medical procedures (Drake et al. 2006; Stoddard et al. 2006), elevated heart rate, pain (Stoddard et al. 2006), quality of family relationships, and parent PTSS (Stoddard et al. 2006; Graf et al. 2011). Risk factors that have been identified for PTSS in parents (mostly of older injured children) include: burn size, guilt, parent–child conflict, peritrauma dissociation, child PTSS (Hall et al. 2006; Bakker et al. 2013), prior trauma exposure, and perceptions of child's pain and life threat (Kassam-Adams et al. 2009).

Although research in this field is growing, further research is needed to understand the impact of childhood trauma on parents of young children, to confirm and identify potential risk factors for persistent child and parent PTSD, and to determine the direction of the relationship between child and parent distress. This information is needed to aid the early identification of young children and parents who are at greater risk of developing persistent adverse outcomes as well as to identify targets for treatment. Early identification of families at risk, coupled with the implementation of effective treatments, is vital to current child and parent functioning, the quality of the parent–child relationship, and, potentially, the child's long-term physical and psychological recovery and developmental trajectories.

Therefore, the first aim of this study was to document the prevalence and prospective change in parent psychological distress (PTSD, depression, anxiety, and stress) at 0–2 weeks, and 1 and 6 months following their young child's burn injury. The second aim of this study was to prospectively examine a set of risk factors for child and parent PTSD as it unfolded over 6 months after burn trauma, in order to develop a model of risk factors for child and parent PTSD. In particular, this study aimed to explore the relationship between child and parent distress over time. It was hypothesized that there would be cross-sectional and longitudinal associations between child and parent PTSS. We also hypothesized that premorbid child psychological functioning and injury severity would be risk factors for child PTSS, and that prior trauma history, injury severity, and guilt would be predictive of parent PTSS. The prevalence and course of the children's posttrauma reactions in this study have been previously described (De Young et al. 2011a, 2012).

Methods

Participants

Data for this article were collected as part of a larger longitudinal research project investigating the psychosocial impact of burn injury in preschoolers and their parents. To date, this cohort of children has been described in two other articles (De Young et al. 2011a, 2012). Participants in the current study were parents of children ages 1–6 years who required medical treatment for an unintentional burn injury. Participants were excluded if 1) the parent's English was insufficient to complete interviews or questionnaires, 2) the injury was a result of suspected child abuse or neglect, and/or 3) the child had a pervasive developmental disorder.

A total of 329 children were eligible for inclusion in the study; however, only 196 families could be approached before the child was discharged from the outpatient unit. One hundred and seventy-seven parents provided informed consent to participate in the study, of which 116 completed the Time 1 (T1) questionnaires within 2 weeks of the child's injury (mean=11.45 days [SD=7.90]). Participants who did not complete T1 questionnaires were retained in the study. At Time 2 (T2), 130 parents completed the interview 1 month post-accident (mean=38.39 days, SD=8.69), and 116 returned the 1 month questionnaires (mean=44.60 days SD=17.16). At Time 3 (T3), 6 months post-injury, 125 parents completed the interview (mean=199.64 days, SD=16.21), and 115 returned questionnaires (mean=210.33 days, SD=34.63).

Measures

Diagnostic Infant Preschool Assessment (DIPA)

The DIPA (Scheeringa and Haslett 2010) is a semistructured diagnostic interview conducted with the primary caregiver of children ages 1–6 years. Overall, the DIPA has demonstrated acceptable test–retest reliability (Scheeringa and Haslett 2010). The PTSD module was used to assess PTSD diagnosis and total PTSS in the child. The proposed American Psychiatric Association, Diagnostic and Statistical Manual of Mental Disorders, 5th edition (DSM-V) (American Psychiatric Association 2013) PTSD criteria for preschool children (De Young et al., 2011a) was the diagnostic classification system used. To check inter-rater reliability, a random sample of 13 interviews was audiotaped and blind-rated by a trained research assistant. The intraclass correlation coefficient for total PTSS was 0.98.

Child Behavior Checklist for ages 1.5-5 (CBCL/1.5-5)

The CBCL/1.5-5 (Achenbach and Rescorla 2000) is a 100 item parent report checklist that measures emotional and behavioral functioning in children ages 1.5–5 years. The CBCL/1.5-5 has demonstrated good psychometric properties (Achenbach and Rescorla 2000). The CBCL was used at T1 to get an indication of the child's preinjury level of functioning. In this study, Cronbach's α was excellent: 0.93 for the Total Problems scale.

Posttraumatic Diagnostic Scale (PDS)

The PDS (Foa 1995) is a 49 item self-report questionnaire that is used to screen and assess for PTSD in adults. The items correspond with the American Psychiatric Association, Diagnostic and Statistical Manual of Mental Disorders, 4th edition (DSM-IV) (American Psychiatric Association 1994) PTSD criteria. The PDS was used to provide a measure of total number of prior trauma exposures, PTSD diagnosis, number of symptoms present, symptom severity rating, and level of impairment in functioning. Parents received a “probable PTSD diagnosis” if their answers met DSM-IV PTSD Criteria A–F. The following manual (Foa, 1995) recommendations for symptom severity rating category cutoffs were used: ≤10=mild, 11–20=moderate, 21–35=moderate to severe, and ≥36=severe. The Impairment in Functioning cutoffs and categories were as follows: 0=no impairment, 1–2=mild, 3–6=moderate, and 7–8=severe. To get an indication of whether parent-reported PTSD symptoms were related specifically to their child's burn injury, parents were asked to complete the questions related to PTSD symptoms and impairment with respect to their child's injury, rather than the traumatic event that bothered them the most as in the original version. Psychometric evaluation has demonstrated acceptable to excellent internal consistency, good test–retest reliability, and acceptable convergent and concurrent validity (Foa et al. 1997). Cronbach's α in the current study was good to excellent: 0.93 at T2 and 0.89 at T3.

Depression Anxiety Stress Scale-21 (DASS-21)

The DASS-21 (Lovibond and Lovibond 1995) is a 21 item self-report measure that was used to assess for depression, anxiety, and stress symptomatology in the parents. Items on the DASS were summed and multiplied by 2 to provide scores on the Total Scale, Depression Scale, Anxiety Scale, and Stress Scale. The cutoffs for severity labels that are recommended in the DASS manual (Lovibond and Lovibond 1995) were also used to report the number of parents with symptoms of moderate severity or greater (Depression=14⁺, Anxiety=10⁺, Stress=19⁺). The DASS-21 has demonstrated acceptable to excellent reliability and convergent and discriminant validity (Lovibond and Lovibond 1995; Henry and Crawford 2005). In this study, across time points, Cronbach's α was good to excellent: 0.93–0.94 for Total Score, 0.85–0.87 for Depression, 0.80–0.86 for Anxiety, and 0.87–0.91 for Stress.

Brief COPE

The Brief COPE (Carver 1997) is a 28-item self-report questionnaire designed to assess coping styles that people use to deal with stressful situations. Respondents are required to rate the extent to which they use each coping style on a four point scale ranging from 1 (“I haven't been doing this at all”) to 4 (“I've been doing this a lot”). The self-blame subscale was used for this study to assess parent guilt. This subscale consists of the following two questions: “I've been criticizing myself” and “I've been blaming myself for things that happened.” In this study, Cronbach's α at 0–2 weeks=0.82, at 1 month=0.78, and at 6 months=0.77.

Demographic, medical, and history information

Demographic information was obtained from questions included in the T1 questionnaire booklet. Injury severity information was obtained from the child's medical record at 6 months post-burn. Burn severity was measured using burn size (i.e., percentage total body surface area [%TBSA]) and number of invasive medical procedures (e.g., dressing changes, debridement, skin grafts).

Procedure

During burns outpatient clinics, a member of the research team screened the clinic attendance list to identify families who met study inclusion criteria. These researchers would then introduce themselves to the family, check eligibility, and explain details of the study. After parents had read the information sheet and signed the consent form they were provided with the T1 questionnaire booklet (CBCL, DASS, Brief COPE) to complete within 2 weeks of the burn injury. Parents were followed up 1 and 6 months later to complete the DIPA over the telephone and to fill in and return the T2 and T3 questionnaire booklets (CBCL, PDS, DASS, Brief COPE). The study was approved and conducted in accordance with the University of Queensland Human Ethics Committee and Royal Children's Hospital Ethics Committee.

Data analysis

Missing data

The data used for this study were taken from 120 participants. Of these cases, 92 had completed data for all outcomes and independent variables on all three occasions, and 28 had some missing data. Cases were only included for analysis if they had at least one record of PTSS for the child and parent over the 6 month period. The only pattern evident was that the mean guilt score at 1 month was statistically significantly higher for 18 parents who did not complete T1 questionnaires (mean=4.6) compared with the rest (mean=3.47), Mann–Whitney U test, Z=−2.6, p=0.01. There were no other patterns of missing data and the data were deemed to be missing completely at random (Little's MCAR χ ²=119.5, df=126, p=0.65).

Statistical method

Descriptive analyses were used to summarize sample characteristics and prevalence rates. Changes in parents' PTSD diagnostic status and symptom levels were examined using McNemar's test of change and paired sample t tests. Risk factor variables for child and/or parent PTSS previously identified in the medical and young child trauma literature were first examined using bivariate correlations. Next, separate hierarchical multiple regression analyses were performed to identify potential risk factors for the four outcome variables (child and parent 1 and 6 month total PTSS). Order of entry of variables in analyses was based on temporal sequence and theoretical relevance, and consisted of the following blocks: 1) Premorbid factors (gender, age, prior trauma history, premorbid emotional and behavioral problems), 2) injury-related factors (%TBSA, number of invasive medical procedures), 3) acute parent distress (total depression, anxiety and stress symptoms, guilt), 4) child and/or parent 1 month PTSS and parent guilt, and 5) child or parent 6 month PTSS and parent guilt. Variables that were not significant at p<0.05 within the block when first entered into the model were excluded and the model was rerun. Goodness of fit of the four final regression models was then assessed. IBM® SPSS® was employed to conduct these analyses.

Finally, path analysis (or structural equation modelling [SEM]) was used to explore and test the complex patterns of relationships between the hypothesized risk factor variables and child and parent PTSS. Unlike regression, in which a dependant variable must be specified, SEM is a more general approach in which a variable can act as both an independent and a dependant variable, and estimation of these dual relationships can be undertaken within the same model. The path analyses are exploratory in nature, in part because of the overall sample size of the study, and are presented to provide a direction for future research.

The initial model incorporated two sets of observed exogenous predictors (i.e., predictors that do not regress on other variables), namely age; premorbid emotional and behavioral problems; and %TBSA for child PTSS; and number of invasive medical procedures; prior trauma history; parent total depression, anxiety, and stress symptoms; and guilt on each occasion for parent PTSS. In order to obtain a parsimonious good fitting model, the modelling process was conducted using a succession of six nested models. Each model was evaluated against its predecessor, and parameters constrained, based on the χ² test for fit, change in fit, examination of parameter estimates, their critical ratios (CRs), CRs for differences between pairs of parameters, p values, and theoretical justifications. Full information maximum likelihood (FIML) was used, as this was appropriate for dealing with missing data. Final model fit was also examined using χ² , the root mean square error of approximation (RMSEA), and comparative fit index (CFI). The χ² statistic provides a test of the null hypothesis that the variance–covariance matrix implied by the model fits the observed one. An RMSEA of <0.05 is recommended for an acceptable model fit, and the CFI should be >0.95 (Schermelleh-Engel et al. 2003). The final model was then re-estimated using Bayesian methodology, which more appropriately models the underlying distribution of the data, rather than relying on the assumptions of normality. IBM® SPSS® Amos 20.0.0 was used for analyzing the paths formulated in the SEM.

Results

Sample characteristics

The sample for this study consisted of 120 parents (111 mothers and 9 fathers). Parents' mean age at time of burn was 32.88 years (SD=5.36, range 17–46), the majority were married (77[68%]), and employed (70[63%]) and 45 (40%) had a university degree or higher. Sixty-one (51%) children were boys and mean age at time of burn was 2.66 years (SD=1.49, range=1–6.69). The most common type of burn was caused by contact with a hot object (50 [42%]), followed by a scald (46 [38%]), fire/flames (12 [10%]), friction (8 [7%]) and chemical/electrical sources (4 [3%]). The mean %TBSA was 3.16 (SD=4.35, range=1–27), children experienced 4.63 invasive procedures on average (SD=2.99, range=1–15) and only 24 (20%) were hospitalized. The accident was witnessed by 99 (83%) of the participating parents. Parents reported that the most traumatic part for them was the actual burn injury (55 [47%]), dressing changes (21 [18%]), both the burn and dressings (17 [15%]), seeing their child in pain (6 [5%]), the hospital experience (6 [5]%), uncertainty (4 [3%]), travel for treatment (4 [3%]), skin graft (2 [2%]), and nothing (1 [1%]). We have reported previously that 25% of the children in this sample at 1 month and 10% at 6 months were diagnosed with PTSD (De Young et al. 2011a). Children had a mean number of 3.12 (SD=2.82, range=0–11) PTSD symptoms at 1 month and 1.55 (SD=1.55, range=0–10) PTSD symptoms at 6 months.

Prevalence and course of parent trauma reactions

Depression, anxiety and stress

During the acute post burn period, nearly 25% of parents experienced moderate to extremely severe levels of depression (25 [22%]), anxiety (21 [18%]), and stress (26 [23%]). After 1 month, 15 (14%), 13 (12%), and 17 (15%) parents were within the moderate to extremely severe range for depression, anxiety, and stress, respectively. At 6 months, the majority of parents (107 [93%], 106 [92%], 106 [92%]) were in the normal to mild range. There was a significant overall reduction in DASS total scores over 6 months (mean [T1]=24.69, SD=19.87 and mean [T3]=11.56, SD=18.15; t [99]=7.17, p<0.0001).

PTSD

Twenty-five (22%) parents had a probable PTSD diagnosis at 1 month post-injury. There was a significant reduction in both PTSD prevalence rates (McNemar test, p<0.001) and PTSS from 1 to 6 months (mean=5.17, SD=4.69 to mean=2.61, SD=3.49, t [109]=7.22, p<0.001). However, 6 (5%) parents continued to have a probable PTSD diagnosis at 6 months, 13 (11%) were experiencing PTSS in the moderate to moderate-severe range, and the symptoms of 15 (13%) parents caused moderate impairment in functioning.

Predictors of child and parent PTSS

The bivariate correlations between variables are presented in Table 1 and the results from the hierarchical multiple regression analyses predicting child and parent PTSS can be seen in Tables 2 and 3. As shown in Table 2, each block of variables contributed significantly to the prediction of child PTSS at 1 month, with the combination of risk factors accounting for 43% of the variance. Greater levels of concurrent parent PTSS followed by higher levels of premorbid emotional and behavioral difficulties and larger %TBSA each accounted for a significant amount of unique variance in child PTSS at 1 month. For 6 month child PTSS, the combination of variables in the final model accounted for 59% of the variance, with significant independent predictors including 1 month child PTSS, parent PTSS at 1 and 6 months, %TBSA, and premorbid psychological functioning.

Table 1.

Bivariate Correlations Among Predictor and Outcome Variables

Variables	1	2	3	4	5	6	7	8	9	10	11	12	13	14
1. Child gender	1
2. Child age	0.05	1
3. Premorbid CBCL Total	−0.01	0.14	1
4. Parent trauma history	0.11	−0.06	0.16	1
5. %TBSA	0.09	0.14	0.10	−0.02	1
6. No0. invasive procedures	0.02	0.02	0.03	−0.05	0.55^***	1
7. T1 parent guilt	−0.15	−0.14	0.12	0.14	−0.10	0.02	1
8. T1 parent distress	−0.06	−0.22^*	0.18	0.12	0.07	0.16	0.52^***	1
9. T2 parent guilt	−0.02	−0.10	0.15	0.19^*	−0.08	0.15	0.66^***	0.41^***	1
10. T2 parent PTSS	0.04	−0.12	0.24^*	0.25^**	0.30^***	0.32^***	0.28^**	0.51^***	0.38^***	1
11. T2 child PTSS	0.07	0.16	0.37^***	0.23^*	0.37^***	0.28^**	0.19	0.35^***	0.28^**	0.60^***	1
12. T3 parent guilt	−0.07	0.12	0.15	0.17	0.02	0.18	0.48^***	0.39^***	0.59^***	0.48^***	0.22^*	1
13. T3 parent PTSS	0.09	−0.03	0.24^*	0.22^*	0.34^***	0.43^***	0.20^*	0.49^***	0.30^**	0.62^***	0.49^***	0.48^***	1
14. T3 child PTSS	0.06	0.11	0.38^***	0.19^*	0.45^***	0.35^***	0.08	0.24^*	0.20^*	0.58^***	0.64^***	0.29^**	0.54^***	1

p<0.05, ^** p<0.01, ^*** p<0.001.

CBCL, Child Behavior Checklist; PTSS, posttraumatic stress symptoms; T1, Time 1 (0–2 weeks); T2, Time 2 (1 month); T3, Time 3 (6 months); %TBSA, percentage total body surface area burned.

Table 2.

Summary of Hierarchical Multiple Regression Analyses for Predicting 1 and 6 Month Child PTSS

		1 month child PTSS (n=98)					6 month child PTSS (n=96)
Block	Risk factor variables	R²	ΔR²	B	SE	ß	R²	ΔR²	B	SE	ß
1	Pretrauma	0.12	0.12^***				0.14	0.14^***
	CBCL total problems			0.04	0.02	0.22^**			0.02	0.01	0.15^*
2	Injury related	0.23	0.11^**				0.29	0.16^***
	%TBSA			0.12	0.05	0.21^*			0.07	0.04	0.15^*
3	Acute parent distress	0.30	0.07^**				0.32	0.02
	T1 DASS Total			0.01	0.01	0.07			−0.02	0.01	−0.16
4	Parent PTSS	0.43	0.12^***				0.47	0.15^***
	T2 Parent PTSS			0.25	0.06	0.43^***			0.10	0.05	0.21^*
5^a	Child PTSS	–	–				0.55	0.08^***
	T2 Child PTSS			–	–	–			0.29	0.08	0.36^***
6^a	Parent PTSS	–	–				0.59	0.04^**
	T3 Parent PTSS			–	–	–			0.16	0.06	0.26^**

6 month blocks.

p<0.05, ^** p<0.01, ^*** p<0.001.

CBCL, Child Behavior Checklist; DASS, Depression Anxiety and Stress Scale; PTSS, posttraumatic stress symptoms; T1, Time 1 (0–2 weeks); T2, Time 2 (1 month); T3, Time 3 (6 months); %TBSA, percentage total body surface area burned.

Table 3.

Summary of Hierarchical Multiple Regression Analyses for Predicting 1 and 6 Month Parent PTSS

		1 month parent PTSS (n=96)					6 month parent PTSS (n=95)
Block	Risk factor variables	R²	ΔR²	B	SE	ß	R²	ΔR²	B	SE	ß
1	Pretrauma	0.07	0.07^**				0.05	0.05^*
	Trauma history			0.43	0.31	0.12			0.22	0.22	0.08
2	Injury related	0.18	0.11^**				0.26	0.21^***
	No. procedures			0.23	0.12	0.15			0.31	0.09	0.27^**
3	Acute parent distress	0.34	0.16^***				0.39	0.13^***
	T1 DASS Total			0.07	0.02	0.28^**			0.05	0.02	0.26^**
4^a	Parent guilt	0.37	0.03^*				–	–
	T2 Parent guilt			0.41	0.27	0.14			–	–	–
5/4	Child PTSS	0.46	0.09^***				0.43	0.04^*
	T2 Child PTSS			0.60	0.15	0.35^***			−0.05	0.13	−0.04
5^b	Parent PTSS	–	–				0.49	0.05^**
	T2 Parent PTSS			–	–	–			0.17	0.08	0.22^*
6^b	Child PTSS	–	–				0.53	0.05^**
	T3 Child PTSS			–	–	–			0.47	0.16	0.30^**

1 month block only; ^bAdditional 6 month blocks.

p<0.05, ^** p<0.01, ^*** p<0.001.

DASS, Depression Anxiety and Stress Scale; PTSS, posttraumatic stress symptoms; T1, Time 1 (0–2 weeks); T2, Time 2 (1 month); T3, Time 3 (6 months).

Each block of variables contributed significantly to the prediction of parent PTSS and together accounted for 46% of the variance in parent PTSS at 1 month, and 53% of the variance at 6 months (Table 3). At the final step, acute parent distress and concurrent child PTSS emerged as the best predictors of parent PTSS at 1 month. The independent predictors for parent PTSS at 6 months were number of invasive procedures, acute parent distress, parent 1 month PTSS, and concurrent child PTSS.

Path analyses

The best fitting model from the exploratory path analyses is presented in Figure 1. Overall, this model appears to fit the observed variance-covariance matrix: χ²=73.03 (60), p=0.121; RMSEA=0.04, 90% CI=0.00–0.07; CFI=0.97. Three pathways had statistically nonsignificant direct effects (p>0.05); however, these were retained in the final model, given that constraining their parameters to zero resulted in a significantly inferior fit to the final model, χ² change=8.06 (3), p=0.045.

FIG. 1.

Path analytic model of child and parent posttraumatic stress at 1 and 6 months. This is the final Bayesian model showing standardized regression pathways of direct effects and squared multiple correlations for the outcomes. Dashed lines depict pathways that are approaching significance. Rectangles=outcome variables; rectangular outward arrows=predictor variables. DASS, Depression Anxiety Stress Scale; PTSS, posttraumatic stress symptoms; %TBSA, percentage total body surface area burned; T1, Time 1; T2, Time 2; T3, Time 3.

Similar to the results reported for the regression analysis, concurrent parent PTSS was a significant predictor of child PTSS at 1 month (β=0.51, p<0.001) and 6 months (β=0.19, p<0.05). The longitudinal relationship identified in regression analysis between parent PTSS at 1 month and child PTSS at 6 months approached statistical significance in the final model (β=0.16, p=0.08). There was a strong indirect effect of 1 month parent distress to 6 month child PTSS via its effect on child 1 month PTSS (β=0.30; 95% CI: 0.15–0.45). Because of the complexity of analyzing concurrent reciprocal, as well as temporal, effects within the same model, only the initial effect of parent PTSS on child PTSS at 1 month was significant. Given that the concurrent effect from child to parent PTSS was not statistically significant at either time point, it was excluded from the model. This resultant model showed significant concurrent effects of parent PTSS on child PTSS at both 1 and 6 months. Greater %TBSA and higher premorbid emotional and behavioral scores were directly associated with higher child PTSS at 1 and 6 months, and there was a positive association between age and child PTSS at 1 month. Child PTSS levels at 1 month were predictive of child PTSS at 6 months. Guilt had a significant direct effect on acute parent distress and 6 month PTSS, greater number of invasive procedures predicted higher parent PTSS at 1 and 6 months, and prior trauma history was associated with more PTSS at 1 month. There were significant positive associations between parent distress variables over time.

Discussion

This study demonstrated that parents of preschool children exposed to traumatic burn injury are at risk of a range of posttrauma reactions, particularly within the first month. Within 2 weeks, 20–23% of parents reported experiencing moderate to extremely severe levels of depression, anxiety, and stress, respectively. These rates were 12–15% at 1 month, and 7–8% at 6 months. Whereas depression and anxiety symptoms were elevated in the acute post-burn period, symptom levels at 6 months were comparable to general population levels in Australia (Andrews et al. 2001). At 1 month post-burn, 22% of parents had a probable PTSD diagnosis. At 6 months, 5% of parents met PTSD criteria and 11% were experiencing symptoms of at least moderate severity. These PTSD rates are comparable to those reported for mothers of burn- injured children (9–18.6%; Bakker et al. 2013) and are higher than what would be expected in the normal population (1.3%; Andrews et al. 2001).

Consistent with hypotheses, parent distress was significantly associated with child PTSS concurrently and longitudinally in the correlation and regression analyses, and concurrently at each time point in the path analyses. Child PTSS was positively associated with parent PTSS at each time point in the bivariate correlations and concurrently but not longitudinally related to parent PTSS in the regression analyses. It may be that the association found between parent and child distress is in part the result of their shared experience of the event, and that they are both reacting in similar ways (e.g., both becoming distressed by reminders of the child's traumatic medical procedures) (Smith et al. 2001). These findings may also be indicative of a shared biological or genetic vulnerability to psychopathology (Saxe et al. 2005; Drury et al. 2013). Further, these findings may be reflecting complex bidirectional relationship interactions (Scheeringa and Zeanah, 2001; Smith et al. 2001). Young children have limited emotional self-regulation capacities, and it is developmentally appropriate to rely on caregivers for assistance with emotion regulation, particularly during times of distress. Elevated distress in parents may compromise their ability to be attuned and emotionally available to assist their children in regulating pain or distress following their trauma (Lieberman 2004; Saxe et al. 2005; Stoddard et al. 2006). In turn, the children's traumatic stress reactions may exacerbate parents' distress or feelings of guilt, and further compromise their ability to support their children over time. Additionally, children, in particular infants and preschoolers, rely on their parent's reactions to determine how to interpret or respond to an event, and may model their fear responses and maladaptive coping mechanisms (Nugent et al. 2007). Furthermore, children and parents may negatively reinforce each other by avoiding reminders or conversations about the burn event (e.g., for fear of upsetting each other), thus impeding their ability to habituate to or process and understand the event (Smith et al. 2001). However, the design of this study does not allow conclusions to be made about such mechanisms.

The exploratory path analyses were conducted in an attempt to elucidate the direction of the relationship between child and parent distress. Because of the complexity of path analyses, in which concurrent pathways are examined within the same model, the lack of a sufficiently large sample size made it difficult to separate the effect of child on parent from that of parent on child. Consequently, only the effect of the parent on the child emerged as significant. However, this finding was similar to those of Landolt and colleagues (2012), who tested the mutual prospective influence of child and parental PTSS following medical trauma in older children (6.5–16 years). Specifically, they found that high PTSS in mothers and fathers at 5–6 weeks was longitudinally related to child PTSS at 12 months, whereas child PTSS was not found to influence parental PTSS concurrently or over time. Together, these findings suggest that the parents' responses to a traumatic event may play a particularly important role in a young child's psychological recovery. This makes sense in the context of early childhood, given that young children are so dependent on the parent–child relationship in almost every way, and have few external influences. In contrast, there are many other factors that may play a more significant role in how adults respond to traumatic events (e.g., social support, other life stressors, coping strategies). However, further research is needed with larger sample sizes to confirm the direction of this relationship between child and parent distress.

Other potential risk factors identified for child PTSS included higher levels of premorbid emotional and behavioral difficulties and larger burn size. Scheeringa and colleagues (2005) have found that the pretrauma externalizing behavior was related to child PTSS, through its interaction with witnessing a threat to their caregiver. Additionally, research has consistently found %TBSA to contribute significantly to the prediction of child PTSS (Drake et al. 2006; Stoddard et al. 2006). There have, however, been inconclusive findings for other measures of injury severity in older children with a range of unintentional injuries (Cox et al. 2008). Size of burn may have a specific influence on PTSS in burn survivors, possibly through its influence on the noradrenergic system (Stoddard et al. 2006), or because scarring can leave a continual visual reminder of the trauma. Age is a variable that is regularly tested and was therefore included as a potential risk factor in all analyses, to allow comparisons with other studies. A positive association between age and 1 month child PTSS was found only in the path analyses. Age is an inconsistent risk factor in the young child trauma literature (Scheeringa et al. 2005, 2006; Feldman and Vengrober 2011).

Consistent with hypotheses and other research (Fukunishi 1998; Bakker et al. 2010, 2012), guilt was identified as a risk factor for parent distress. The relationship between parents' prior trauma exposure and their PTSS is also consistent with previous research (Kassam-Adams et al. 2009). Parents who have experienced prior traumas may be more vulnerable to subsequent stressful life events. Alternatively, PTSS levels 1 month post-accident may partly reflect ongoing symptomatology or an exacerbation of prior PTSS resulting from pre-existing traumas. The number of invasive procedures experienced by the child was a significant predictor of parent PTSS. There are many potentially traumatic aspects associated with medical procedures (e.g., witnessing child's pain, participation in the child's restraint). These can be extremely stressful for parents and may trigger or intensify their own traumatic stress symptoms (Stoddard et al. 2006) or exacerbate feelings of guilt. Indeed, 45% of parents reported that the most traumatic part of the burn event was related to their child's medical treatment and hospital experiences. Furthermore, consistent with other research, parents' initial distress levels predicted their PTSS over time (Kassam-Adams et al. 2009; Landolt et al. 2012).

Limitations and future directions

Questionnaires rather than structured diagnostic interviews were used to assess parents, which may have resulted in less accurate and sensitive indicators of parental distress. Guilt was also only assessed using two questions, which were asked within the context of how people were coping with the stress in their lives, and not specifically about parents' self-blame in relation to their children's burn accident. Therefore, it is possible that this did not provide the most accurate measure of self-blame about the accident. Furthermore, because of child age, parents reported on both their own and their child's functioning. Research has indicated that parent report has a tendency to underestimate internalizing symptoms in children (Scheeringa et al. 2006; Meiser-Stedman et al., 2008). Additionally, parent's psychological health may have influenced ratings of their child's distress and may explain the unidirectional relationship found between child and parent PTSS in the path analysis (Smith et al. 2001). However, because of the limited communication abilities of preschool children, this issue is difficult to avoid, and represents a common problem faced with all research conducted with this population. To assist with minimizing bias, parents were required to give detailed examples of their child's symptoms before they were counted. Additionally, because of the small sample size (which can affect the power and reliability of the model), the results from the path analysis need to be treated as exploratory, and they require replication. Moreover, data were collected for one parent, of whom the majority were mothers; therefore, these findings may not generalize to fathers. Other considerations are that the study findings may not generalize to young children with other injury types, different trauma populations, those from lower socioeconomic backgrounds, and ethnic minorities.

This study has a number of strengths and makes an important contribution to the young child PTSD literature, in which investigations of parent distress and risk factors for young child PTSS are still relatively sparse. However, further longitudinal research with larger sample sizes is needed to further examine the direction of the relationship between parent and child distress and confirm the presence of these risk factors. Directions for future research include utilization of diagnostic interviews for parent distress, using a measure that specifically assesses parental guilt in relation to the child's trauma, including multiple informants (e.g., fathers, other caregivers) and examination of risk factors for longer periods, in various trauma populations. Additionally, it would be beneficial to investigate other potential predictors and mediators of child and parent distress, including physiological reactivity (e.g., heart rate, cortisol), pain, parenting style, and attachment.

Conclusions

Parents of young children are at risk of developing PTSD, anxiety, and depression following their child's trauma. Significant positive associations were found between child and parent PTSS. The exploratory findings from this study suggest that parents' responses to a traumatic event may play a particularly important role in a young child's psychological recovery. However, further research is needed to confirm the direction of the relationship between child and parent distress. Other risk factors for child PTSS include premorbid emotional and behavioral difficulties and larger burn size. Other parent PTSS risk factors were found to be prior trauma history, acute distress, greater number of child invasive procedures, and guilt. This study has identified variables that could be incorporated into screening tools or targeted by early intervention protocols to prevent the development of persistent child and parent PTSS following medical trauma.

Clinical Significance

This study has important clinical implications for the early evaluation and treatment of young children following injury. First, these findings highlight that health practitioners need to be aware that parents are also at risk of developing serious psychological disorders, in particular PTSD, following their child's injury. It is therefore important to minimize the potential trauma associated with medical procedures wherever possible (e.g., through adequate procedural preparation). From an assessment perspective, it is important that a brief, cost-effective, and reliable screening instrument be developed for young children, to evaluate whether it can be routinely administered in hospitals, to identify and monitor children at high risk of developing persistent PTSS. Furthermore, the association evident between parent and child distress highlights the importance of screening and monitoring parents at risk of developing PTSS over the course of their child's treatment and physical recovery. The results from this study suggest that any intervention to reduce child PTSS needs to be undertaken within the context of the parent–child relationship. Intervention for families identified as at risk may include 1) treating parent's PTSD symptoms or guilt responses, 2) providing parents and child (where developmentally appropriate) with coping strategies to address the child's trauma responses, and 3) addressing any parent–child relationship difficulties.

Footnotes

Acknowledgments

We acknowledge and thank Ms. Jacelle Lang and Dr. Robyne Le Brocque for helpful feedback on this manuscript, and we also thank the Royal Children's Hospital staff and the children's parents for their participation in this research project.

Disclosures

No competing financial interests exist.

References

Achenbach

, Rescorla

: Manual for the ASEBA Preschool Forms & Profiles. Burlington, VT: University of Vermont, Research Center for Children, Youth, & Families; 2000.

American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, 4th ed. Washington, DC: American Psychiatric Association; 1994.

American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, 5th ed. Washington, DC: American Psychiatric Association; 2013.

Andrews

, Henderson

, Hall

: Prevalence, comorbidity, disability and service utilisation: Overview of the Australian National Mental Health Survey. Br J Psychiatry, 178:145–153, 2001.

Australian Institute of Health and Welfare: A Picture of Australia's Children 2009. Cat. no. PHE 112. Canberra: Australian Institute of Health and Welfare; 2009.

Bakker

, Maertens

KJP

, Van Son

MJM

, Van Loey

NEE

: Psychological consequences of pediatric burns from a child and family perspective: A review of the empirical literature. Clin Psychol Rev, 33:361–371, 2013.

Bakker

, Van Loey

, Van Son

, Van der Heijden

: Brief report: Mothers' long-term posttraumatic stress symptoms following a burn event of their child. J Pediatr Psychol, 35:656–661, 2010.

Bakker

, Van Loey

NEE

, Van der Heijden

PGM

, Van Son

MJM

: Acute stress reactions in couples after a burn event to their young child. J Pediatr Psychol, 37:1127–1135, 2012.

Carpenter

, Stacks

: Developmental effects of exposure to intimate partner violence in early childhood: A review of the literature. Child Youth Serv Rev, 31:831–839, 2009.

10.

Carver

: You want to measure coping but your protocol's too long: Consider the Brief COPE. Int J Behav Med, 4:92–100, 1997.

11.

Cox

, Kenardy

, Hendrikz

: A meta-analysis of risk factors that predict psychopathology following accidental trauma. J Spec Pediatr Nurs, 13:98–110, 2008.

12.

De Young

, Kenardy

, Cobham

: Diagnosis of posttraumatic stress disorder in preschool children. J Clin Child Adolesc Psychol, 40:375–384, 2011a.

13.

De Young

, Kenardy

, Cobham

: Trauma in early childhood: A neglected population. Clin Child Fam Psychol Rev, 14:231–250, 2011b.

14.

De Young

, Kenardy

, Cobham

, Kimble

: Prevalence, comorbidity and course of trauma reactions in young burn-injured children. J Child Psychol Psychiatry, 53:56–63, 2012.

15.

Drake

, Stoddard

, Murphy

, Ronfeldt

, Snidman

, Kagan

, Saxe

, Sheridan

: Trauma severity influences acute stress in young burned children. J Burn Care Res, 27:174–182, 2006.

16.

Drury

, Brett

, Henry

, Scheeringa

: The association of a novel haplotype in the dopamine transporter with preschool age posttraumatic stress disorder. J Child Adolesc Psychopharmacol, 23:236–243, 2013.

17.

Feldman

, Vengrober

: Posttraumatic stress disorder in infants and young children exposed to war-related trauma. J Am Acad Child Adolesc Psychiatry, 50:645–658, 2011.

18.

Foa

: Posttraumatic Stress Diagnostic Scale: Manual. Minneapolis: Pearson; 1995.

19.

Foa

, Cashman

, Jaycox

, Perry

: The validation of a self-report measure of posttraumatic stress disorder: The Posttraumatic Diagnostic Scale. Psychol Assess, 9:445–451, 1997.

20.

Fukunishi

: Posttraumatic stress symptoms and depression in mothers of children with severe burn injuries. Psychol Rep, 83:331–335, 1998.

21.

Graf

, Schiestl

, Landolt

: Posttraumatic stress and behavior problems in infants and toddlers with burns. J Pediatr Psychol, 36:923–931, 2011.

22.

Hall

, Saxe

, Stoddard

, Kaplow

, Koenen

, Chawla

, Lopez

, King

: Posttraumatic stress symptoms in parents of children with acute burns. J Pediatr Psychol, 31:403–412, 2006.

23.

Henry

, Crawford

: The short-form version of the Depression Anxiety Stress Scales (DASS-21): Construct validity and normative data in a large non-clinical sample. Br J Clin Psychol, 44:227–239, 2005.

24.

Kassam–Adams

, Fleisher

, Winston

: Acute stress disorder and posttraumatic stress disorder in parents of injured children. J Trauma Stress, 22:294–302, 2009.

25.

Landolt

, Ystrom

, Sennhauser

, Gnehm

, Vollrath

: The mutual prospective influence of child and parental post-traumatic stress symptoms in pediatric patients. J Child Psychol Psychiatry, 53:767–774, 2012.

26.

Lieberman

: Traumatic stress and quality of attachment: Reality and internalization in disorders of infant mental health. Infant Ment Health J, 25:336–351, 2004.

27.

Lovibond

, Lovibond

: Manuscript for the Depression Anxiety Stress Scales, 2nd ed. Sydney: The Psychology Foundation of Australia Inc.; 1995.

28.

Meiser–Stedman

, Smith

, Glucksman

, Yule

, Dalgleish

: The posttraumatic stress disorder diagnosis in preschool- and elementary school-age children exposed to motor vehicle accidents. Am. J. Psychiatry, 165:1326–1337, 2008.

29.

Nugent

, Ostrowski

, Christopher

, Delahanty

: Parental posttraumatic stress symptoms as a moderator of child's acute biological response and subsequent posttraumatic stress symptoms in pediatric injury patients. J Pediatr Psychol, 32:309–318, 2007.

30.

Perry

, Pollard

, Blakley

, Baker

, Vigilante

: Childhood trauma, the neurobiology of adaptation, and “use-dependent” development of the brain: How “states” become “traits.”. Infant Ment Health J, 16:271–291, 1995.

31.

Saxe

, Stoddard

, Chawla

, Lopez

, Hall

, Sheridan

, King

: Risk factors for acute stress disorder in children with burns. J Trauma Dissociation, 6:37–49, 2005.

32.

Scheeringa

, Haslett

: The reliability and criterion validity of the Diagnostic Infant and Preschool Assessment: A new diagnostic instrument for young children. Child Psychiatry Hum Dev, 41:299–312, 2010.

33.

Scheeringa

, Wright

, Hunt

, Zeanah

: Factors affecting the diagnosis and prediction of PTSD symptomatology in children and adolescents. Am J Psychiatry, 163:644–651, 2006.

34.

Scheeringa

, Zeanah

: A relational perspective on PTSD in early childhood. J Trauma Stress, 14:799–815, 2001.

35.

Scheeringa

, Zeanah

: Reconsideration of harm's way: Onsets and comorbidity patterns of disorders in preschool children and their caregivers following Hurricane Katrina. J Clin Child Adolesc Psychol, 37:508–518, 2008.

36.

Scheeringa

, Zeanah

, Myers

, Putnam

: Predictive validity in a prospective follow-up of PTSD in preschool children. J Am Acad Child Adolesc Psychiatry, 44:899–906, 2005.

37.

Schermelleh–Engel

, Moosbrugger

, Müller

: Evaluating the fit of structural equation models: Tests of significance and descriptive goodness-of-fit measures. Methods of Psychological Research Online, 8:23–74, 2003.

38.

Smith

, Perrin

, Yule

, Rabe–Hesketh

: War exposure and maternal reactions in the psychological adjustment of children from Bosnia–Hercegovina. J Child Psychol Psychiatry, 42:395–404, 2001.

39.

Stoddard

, Saxe

, Ronfeldt

, Drake

, Burns

, Edgren

, Sheridan

: Acute stress symptoms in young children with burns. J Am Acad Child Adolesc Psychiatry, 45:87–93, 2006.

40.

United States Department of Health and Services, 2009: Child Maltreatment 2007. Available at http://www.acf.hhs.gov/programs/cb/stats_research/index.htm#can. Accessed 20 January 2014 .