Adolescent cancer survivors’ posttraumatic stress symptoms: Concordance between self-report and maternal-proxy report

Abstract

Subsyndromal posttraumatic stress among pediatric cancer survivors has been associated with negative physical and mental health outcomes. However, adolescent self-report and mother-proxy report of adolescents’ posttraumatic stress symptoms evidenced varying concordance depending on methodology. There was moderate concordance, particularly among younger respondents, when total posttraumatic stress symptoms were viewed continuously and low-moderate concordance when viewed categorically; moderate-strong concordance for only one posttraumatic stress disorder symptom cluster; low-moderate agreement for high-frequency items; and no concordance for identifying caseness. Although a significant subset of pediatric cancer survivors experience posttraumatic stress, mothers and adolescents demonstrate limited symptom, categorical, and caseness agreement, potentially impacting adolescents’ healthcare service utilization.

Keywords

adolescence cancer concordance posttraumatic stress survivorship

Medical advancements in pediatric cancer treatment have led to improved survival rates. As a result, the research focus has shifted to include the long-term psychological effects following intensive treatment. Although many survivors report positive growth related to their cancer experience (Arpawong et al., 2013), a subset of this population experiences intense anxiety and symptoms of posttraumatic stress (PTS; Schwartz and Drotar, 2006). Many researchers have adopted a PTS model to understand pediatric cancer survivors’ reactions following treatment (e.g. Schwartz and Drotar, 2006; Taïeb et al., 2003). This model has received mixed support (Stuber et al., 1998), with the prevalence of diagnosable posttraumatic stress disorder (PTSD) varying widely between studies (2%–20%; Taïeb et al., 2003), which is nearly indistinguishable from rates in the general population (APA, 2000) but higher than the child/adolescent population (Bruce, 2006). Erickson and Steiner (2001) reported a current PTSD prevalence rate of 10 percent in their sample of adolescent and adult survivors of pediatric cancer, in line with estimates; however, 78 percent of survivors endorsed subsyndromal PTS symptoms. In fact, compared to healthy controls, adolescent cancer survivors have been found to differ in frequency of PTSD symptoms but not clinical PTSD diagnosis, with reported PTSD symptoms associated with higher frequency of medical late effects and lower reports of social desirability (Brown et al., 2003). Thus, diagnostic categorization may only capture a subset of the pediatric cancer survivor population experiencing significant PTS symptoms.

PTS has implications for both physical and mental health. Children and adolescents exposed to other forms of trauma (e.g. sexual or physical abuse; involvement in motor vehicle accidents) have higher rates of substance use (Goldstein et al., 2011) and poorer health-related quality of life (Landolt et al., 2009). In adults, PTS symptoms are linked to greater functional impairment, higher rates of comorbid psychiatric diagnoses, and increased suicidal ideation (Marshall et al., 2001). Drawing upon these findings, Lee et al. (2009) proposed a model for pediatric cancer survivors linking health-related uncertainty, PTS, and health behaviors. The authors posit that adolescent cancer survivors live with chronic and persistent uncertainty related to the possibility of developing late effects related to their treatment. While the majority of survivors adopt a healthier lifestyle, a subset of these individuals may develop symptoms of PTS that could lead to increased engagement in health-risk behaviors and failure to engage in positive health behaviors (Lee et al., 2009). Although the model has not yet been formally tested, it suggests that more research is necessary in evaluating the clinical significance of PTS.

As a subset of childhood cancer survivors appears to be at risk for experiencing significant PTS symptoms, of crucial importance is the identification of these symptoms, which often relies on parents to recognize significant levels of distress in their children. Only two studies have investigated the concordance between parent and child report of the child’s symptoms within pediatric cancer survivors (Clawson et al., 2013; Phipps et al., 2005). Each study compared self-report and parent report on the University of California at Los Angeles Posttraumatic Stress Disorder Reaction Index—Revision 2 (UCLA PTSD-RI; Steinberg et al., 2004) in a group of children aged 7–18 years. No group differences were observed on total symptom scores, and parent–child agreement was estimated in the moderate to high range (intraclass correlation (ICC) = .61 for Clawson et al., 2013; r = .69 for Phipps et al., 2005). The authors concluded that the moderate-high level of agreement indicates that parent-proxy report of child symptoms in this population may be adequate in assessing levels of PTS symptoms (Phipps et al., 2005). However, concordance for PTSD caseness was poor (κ = .26; Clawson et al., 2013), and symptom-level agreement was not investigated as a potential explanation. This is an important distinction, as symptom-level agreement has been found to be higher than diagnostic agreement in other childhood anxiety disorders (Comer and Kendall, 2004).

Beyond the pediatric cancer survivor literature, only a few studies have directly investigated parent and child concordance of child-reported PTS symptoms within other pediatric populations. With pediatric survivors of single-event trauma, such as assault or a motor vehicle accident (Meiser-Stedman et al., 2007), children were more likely to meet diagnostic criteria for PTSD based on their own report of symptoms. Agreement for a PTSD diagnosis based on structured clinician administered interviews with the children and parents was merely fair, indicating that parents were not consistently able to identify symptoms of distress reported by their children. For adolescents experiencing a life-threatening injury, there was moderate parent–child concordance at 5 and 12 months post-injury on self-report and parent-proxy measures of PTS symptoms (κ = .56 at 5 months, κ = .51 at 12 months; Ghesquiere et al., 2008).

Within the limited child–parent concordance literature, the child’s age is one important developmental consideration that may impact overall concordance levels of children’s reported symptoms. Studies addressing PTS symptomatology have not yet explored the effect of age (e.g. Clawson et al., 2013; Ghesquiere et al., 2008; Phipps et al., 2005), whereas age-related findings across other psychiatric diagnoses and symptoms are mixed. An early meta-analysis suggested that parents report children’s emotional problems and psychiatric symptoms in a more concordant manner when their child is younger (less than 12 years) than when their child is an adolescent (Achenbach et al., 1987). Overall parent–child agreement in this meta-analysis was significant but low (r = .22), providing support for the importance of incorporating multiple informants when assessing children and adolescents. More recent studies have failed to show this age trend, reporting that age is not a significant factor when examining concordance of child and parent report of children’s anxiety disorder symptoms (e.g. Choudhury et al., 2003). However, inconsistent findings may be due to differences in selected age ranges and the method of including age as a covariate across studies (De Los Reyes and Kazdin, 2005).

This study aims to further elucidate the concordance between adolescent cancer survivors’ self-reported PTS symptoms and their mothers’ report of their children’s PTS symptoms. In line with analyses conducted by Phipps et al. (2005) and Clawson et al. (2013), we investigated mean symptom differences across PTSD total and criteria scores between adolescents and their mothers and examined the degree of association between respondents. In order to describe and explore the clinical implications of mother–adolescent concordance, additional analyses were conducted to gain understanding of the degree of categorical concordance, the impact of child age on the degree of association between child and mother respondents, and the number of survivors who qualified for PTSD caseness by the two sets of respondents. Our study aims to broaden the understanding of PTS in this population by including only adolescents, including age as a moderating variable, and investigating concordance at the item and symptom criteria levels in addition to concordance at the total symptom and caseness levels.

Methods

Participants

Participants were 26 adolescent cancer survivors and 26 mothers (due to inconsistent attendance of fathers at clinic visits). Adolescents ranged in age from 12 to 18 years (mean age = 15.31 years, standard deviation (SD) = 1.82). Consecutive pediatric oncology survivors in a Young Survivors Clinic were recruited for this study. Cancer survivors were eligible for this study if they had completed cancer treatment at least 1 year prior to recruitment (a requirement for inclusion in the survivors clinic; mean time since diagnosis = 9.07 years, SD = 4.31) and if they were not brain tumor survivors (due to concerns about variable cognitive impairment affecting participation). Diagnoses included leukemias (n = 8), lymphomas (n = 6), solid tumors (n = 7), sarcomas (n = 2), and other cancers (n = 3). Mean maternal education was 1–2 years of college, and mean family income was between US$40,000 and US$60,000. All eligible survivors were recruited, and, based on demographic characteristics (Table 1), the sample appears representative of the larger long-term survivor clinic population in the Young Survivors Clinic. The adolescent ethnic composition was 42 percent Hispanic (n = 11), 42 percent non-Hispanic White (n = 11), 8 percent Native American (n = 2), 4 percent Hispanic/Native American (n = 1), and 4 percent other (n = 1).

Table 1.

Sample characteristics of adolescent cancer survivors.

	N
Total sample	26
Male	15 (58%)
Female	11 (42%)
Hispanic	11 (42%)
Caucasian	11 (42%)
Native American	2 (8%)
Hispanic/Native American	1 (4%)
Other	1 (4%)
	Mean (SD)
Age	15.31 (1.78)
Age at diagnosis	6.04 (4.37)
Years since diagnosis	9.27 (4.33)
PTSD-RI: adolescent self-report
B criterion	3.46 (4.62)
C criterion	6.73 (7.50)
D criterion	7.00 (4.10)
Total	17.19 (13.72)
PTSD-RI: parent-proxy report
B criterion	2.46 (2.58)
C criterion	6.31 (6.99)
D criterion	5.31 (4.15)
Total	14.08 (12.44)

SD: standard deviation; PTSD-RI: Posttraumatic Stress Disorder Reaction Index—Revision 2.

Measures

UCLA PTSD-RI

The PTSD-RI is a self-report measure used to assess the severity of Diagnostic and Statistical Manual of Mental Disorders’ (4th ed.; DSM-IV-TR) symptoms for PTSD (American Psychiatric Association, 2000; Steinberg et al., 2004). In this study, a 20-item total score was employed that assesses symptoms across three clusters (i.e. criteria B, C, and D corresponding to re-experiencing, avoidance, and hyperarousal) in children via self-report and via parent-proxy report, with higher scores indicating more severe symptomatology. Items are rated on a five-point frequency scale, with choices ranging from “none of the time” (0) to “most of the time” (4 = almost every day), allowing for a total possible score of 80. The overall score can be used to classify the severity of respondents’ symptomatology (⩽11 = no significant symptoms endorsed, 12–24 = mild, 25–39 = moderate, ⩾40 = severe; Brown et al., 2003). Internal consistency for the PTSD-RI total score in our sample was α = .90 for adolescents and α = .84 for maternal proxies.

Procedure

Adolescent cancer survivors and their mothers were recruited consecutively from a long-term oncology survivor clinic at the University of New Mexico Hospital by a trained clinical psychology graduate student. In order to attend the clinic, patients had to be at least 1 year post-treatment (i.e. 1 year after completion of the maintenance phase). Prior to approaching families, the study was introduced to clinic participants by a pediatric oncologist or a clinic nurse. Potential participants were informed that participating in the study was voluntary and that declining to participate would not affect healthcare service access. A total of 29 adolescents enrolled in the study. Of the four enrolled participants aged 18 years, only one consented and participated in the study without the presence of a parent or guardian. Given the lack of maternal-proxy data on the PTSD-RI, this participant was excluded. Two additional participants’ mothers did not fully complete the PTSD-RI, so these participants were also excluded. In all, 26 adolescents and their mothers were included in the present analyses.

Once informed consent and assent were obtained, a trained research assistant administered the self-report measure (PTSD-RI) of current PTS symptoms in interview format to the adolescents in order to ensure comprehension. Mothers independently completed a parent-proxy version of the measure, reporting on their impressions of their child’s PTS symptoms. Both respondents were instructed to complete the PTSD-RI measures in relation to the adolescents’ “cancer experience.” This study was approved by the Institutional Review Board by the University of New Mexico Health Sciences Center.

Statistical analyses

Descriptive statistics were calculated for all variables. A Kendall Tau correlation was computed to determine ordinal categorical concordance. Pearson-product moment correlations were calculated to investigate concordance of continuous variables (total and symptom cluster PTSD-RI scores). Paired-sample t-tests were employed to determine mother–child respondent mean differences.

Results

Descriptive statistics for PTSD-RI adolescent self-report and maternal-proxy scores are presented in Table 1. Based on adolescent self-report, 12 adolescents reported minimal PTS symptoms, 8 reported symptoms in the mild range, 5 reported moderate symptomatology, and 1 reported symptoms in the severe range. Based on mother-proxy report, 13 adolescents fell in the minimal symptom category, 9 displayed symptoms in the mild range, 3 were moderate, and 1 demonstrated severe symptomatology. Although the distribution of categories appears to be remarkably similar, only 12 of the 26 mother–child dyads agreed on symptom classification. Examining concordance based on the ordinal categorization completed by mother and child, the Kendall Tau correlation coefficient was .38 (p < .05). In order to further examine the concordance of PTS symptoms reported by adolescents and their mothers, a Pearson correlation was conducted between the PTSD-RI adolescent self-report total score and the mother-proxy total score. Adolescent self-reports and mother-proxy measures were significantly correlated (r = .56, p < .01). To further investigate concordance by symptom clusters, correlations were conducted between mother and child on symptom criteria B, C, and D (re-experiencing, avoidance, and hyperarousal, respectively; Figure 1). Only criterion C reached significance, r = .66, p < .001 (criterion B: r = .36, p = .07; criterion D: r = .38, p = .05).

Figure 1.

Correlations between adolescents’ self-report and maternal-proxy report of PTS symptoms (by total symptoms and symptom cluster).

Given interest in evaluating the relationship between adolescent age and concordance, the sample was divided using a median split methodology into a younger group (12–15 years; n = 12) and an older group (16–18 years; n = 14). A Pearson correlation between the PTSD-RI self-report and mother-proxy total scores was conducted for each age group. The concordance between adolescent and maternal report remained significant for the younger group (r = .75, p < .01) but only approached significance for the older group (r = .53, p > .05). To ensure that this difference was related to adolescent age and not time since diagnosis, time since diagnosis was partialed out, and partial correlation results remained similar for both younger adolescents and older adolescents.

In addition to investigating associations between adolescent and maternal report, we were also interested in determining whether the overall magnitude of PTS symptoms was similar for adolescent self-report and mother-proxy measures. A paired-sample t-test between adolescent and maternal reports was conducted to determine whether the mean responses on the PTSD-RI differed between adolescents and mothers. This test was not significant, although the mean adolescent self-report score was slightly higher than the maternal-proxy report (17.19 vs 14.08; both in the mild range). Post hoc exploratory analyses revealed similar patterns across B, C, and D criteria and for older and younger adolescents.

Qualitatively, we were interested in exploring concordance at the item level for this measure, particularly on items that were rated with a frequency of 2 (1–2 times a week) or higher (3 = 2–3 times a week; 4 = every day). For each item that was rated a 2 or higher by either respondent, we evaluated whether the paired respondent rated the item with a similar score (±1). Out of 520 possible paired responses across the 20 items, adolescents or mothers reported a 2 or higher on 161 items (31%). Of these 161 high frequency responses, the adolescent reported higher frequency of symptoms (2 or more points greater) on 63 items (39% of adolescent-identified high-frequency items; 12% of the 520 possible responses), the mother reported higher frequency of symptoms on 35 items (22% of mother-identified high-frequency items; 7% of the 520 possible), and, most importantly, the pairs agreed on reported frequency (within one point) for 63 items (39% of adolescent- and mother-identified high-frequency items; 12% of the 520 possible).

Finally, caseness was determined by investigating the number of adolescent and maternal reports that exceeded PTSD-RI clinical cut-off scores. The commonly used caseness score of 38 was used to classify adolescents as exhibiting a clinically significant level of PTS symptoms (Steinberg et al., 2004). Based on adolescent self-report, two adolescents scored 38 or higher on this measure. Both adolescents fell in the older group, and their mothers did not report scores that reached the 38-point cut-off. For one dyad, the adolescent reported a total score of 39, while their mother only reported a score of 10. For the second dyad, the adolescent reported a total score of 58, while their mother reported 27. Based on the mother-proxy report, only one adolescent qualified for caseness. This adolescent was in the younger group and missed the self-reported cut-off score by one point, indicating a moderate categorical level of self-reported PTS symptoms (and quite similar to the mother-proxy severe categorical report).

Discussion

This study sought to provide a more nuanced understanding of adolescent and maternal concordance on adolescent cancer survivors’ experience of PTS. To begin, we found that the majority of our sample reported subsyndromal PTS symptoms, as evidenced by 14 of our 26 adolescents self-reporting mild to severe PTS symptoms. When compared to other samples of adolescent pediatric cancer survivors (i.e. Brown et al., 2003; Clawson et al., 2013; Kazak et al., 2004), our sample reported a similar mean level of PTS. Furthermore, mean PTS symptomatology in our sample was greater than seen in samples of healthy controls (i.e. Brown et al., 2003). Overall, the results of this study found varying levels of concordance between adolescent self-report and maternal report of PTS symptoms in pediatric cancer survivors. Concordance agreement depended on a host of factors, including whether total PTS symptoms were treated as a continuous variable or a categorical variable, age was included as a covariate, symptom clusters or individual items were investigated, or a clinical cut-off was employed.

Specifically, we found moderate concordance, particularly among younger respondents and their mothers, when total PTS symptoms were treated as a continuous variable and low-moderate concordance when total PTS symptoms were treated as a categorical variable. Moderate-strong concordance was only found for one PTSD symptom cluster (C), with low-moderate agreement for high-frequency items, and no concordance for identifying caseness. This finding is similar to that reported by Clawson et al. (2013; ICC = .61) and Phipps et al. (2005; r = .69). Relatively slight differences in rates of concordance between the studies may be related, at least in part, to study methodology. Our study only included adolescents who were at least 1 year post-treatment, whereas the two prior studies included participants at varying times since diagnosis, including a group of newly diagnosed children and adolescents, with a significant portion of participants still receiving treatment at the time of data collection. While children are undergoing treatment, they typically spend a significant amount of time with their parents. This increased contact could lead to increased similarity in ratings of PTS symptoms, potentially resulting in a higher overall concordance rate. Compared with other types of trauma, our moderate parent–child concordance of PTS symptoms is better than the fair levels of concordance reported in studies assessing PTS after a single-event trauma with a structured clinical interview (i.e. Meiser-Stedman et al., 2007) and similar to studies using self-report or parent-proxy measures after an adolescent injury (i.e. κ values ranging from .51 to .56 in Ghesquiere et al., 2008).

Importantly, our concordance was more pronounced in the younger adolescent group (12–15 years) than in the older adolescent group (16–18 years). Although prior researchers have failed to examine age when considering parent–child agreement on measures of PTS symptoms, this finding is to be expected given developmental considerations. De Los Reyes and Kazdin (2005) emphasize the developmental importance of age in the context of the parent–child relationship; as a child progresses through adolescence, increased autonomy is attained, and changes in child–parent relationships likely accompany the transition from early adolescence (age: 12–15 years) to middle adolescence (age: 16–19 years; De Goede et al., 2009). This developmental transition may, in turn, predict differential concordance rates of children’s emotional and behavioral functioning. Mothers of younger adolescent survivors may be more aware of the level of distress experienced by their children, while older adolescent survivors may be more at risk for experiencing unrecognized distress, given greater autonomy. Such an interpretation is only one possibility, as the results of this study do not provide information regarding the accuracy of reported symptoms or the mechanisms underlying concordance and discordance. Future studies might further examine this discrepancy and compare the “accuracy” of maternal report versus self-report by employing a clinically administered interview and other indices of PTS-related distress. Additionally, small sample size prevented analysis of ethnic group differences, which is another potential avenue for future research, as cultural differences could impact adolescents’ expression of and communication about PTS symptoms, and subsequently, PTS symptom concordance.

Furthermore, concordance in this study varied when PTS symptoms were treated categorically by criterion. Specifically, the concordance between adolescent and mother-proxy report was highest (moderate-strong) for criterion C (avoidance) and only low-moderate for criteria B (re-experiencing) and D (hyperarousal). Our finding may be explained by the nature of the C symptom cluster, wherein criterion C encompasses avoidance behaviors that are often observable and demonstrable, such as restricted range of affect or avoiding activities, places or people that remind them of the trauma, perhaps making them easier for adolescents to admit and parents to observe and report. Parent–child agreement on observable symptoms has often been shown to be stronger than agreement on unobservable symptoms (e.g. Comer and Kendall, 2004). In contrast, parents may be less able to report on their children’s more internal symptoms like nightmares and intrusive recollections (criterion B) and concentration difficulties or sleep problems (criterion D). In addition to being less observable, internalizing symptoms may be less aversive to parents, as they do not interfere with family routines or require behavior management, resulting in poorer identification and lower concordance (Martin et al., 2004). An additional factor that may influence cluster concordance involves the number of PTSD-RI items that address each symptom criterion. For criterion C, adolescents and parents report on 9 items, compared to 5 items in criterion B and 6 items in criterion D. Although the number of items per criterion did not impact the overall cluster means, the dyads had more opportunities for agreement in criterion C.

While the DSM-IV-based measure we used included three major symptom clusters, Diagnostic and Statistical Manual of Mental Disorders (5th ed.; DSM-5) includes four symptom clusters, with avoidance/numbing (criterion C) divided into two distinct clusters: avoidance and persistent negative alteration in cognitions and mood (American Psychiatric Association, 2013). The latter category, in addition to retaining much of the DSM-IV numbing symptoms, also includes new symptoms such as persistent negative emotional states. The final DSM-IV cluster, hyperarousal (criterion D), now includes irritable or aggressive behavior and reckless or self-destructive behavior in addition to alterations in arousal and reactivity present in DSM-IV. Thus, this study’s findings may be somewhat modified if updated DSM-5 criteria were utilized. Nonetheless, because DSM-IV PTSD symptoms are almost universally retained in DSM-5, with changes representing the inclusion of additional criteria, our results appear representative of core PTSD clusters and symptom levels.

At the high-frequency item level, concordance was only low-moderate among paired respondents, suggesting that the items identified by either the adolescent or his/her mother as frequently occurring or being present (at least weekly) for the adolescent were not the same items the other respondent identified as high-frequency items. Thus, the items endorsed at higher levels were largely dissimilar for the two groups, suggesting that the moderate overall concordance found in total scores is driven by different symptom endorsement at the item level. Whether this item-level discrepancy is clinically meaningful is an empirical question for future research, as we could not find another study investigating PTS symptom concordance at the item level.

Of particular clinical significance, two older adolescents qualified for clinical caseness based on their self-report (i.e. exceeding PTSD-RI clinical cut-off). However, their mothers did not identify either of these adolescents as experiencing clinically significant symptomatology (one mother reported no significant symptoms, the other moderate symptoms). From the maternal perspective, one mother identified her younger adolescent as qualifying for caseness, but this child (marginally) did not identify herself/himself. Even lower than the limited concordance reported by Clawson et al. (2013) using the same measure (κ = .26), our caseness finding provides further support for very limited, even concerning, levels of concordance between adolescent and maternal report when identifying individuals who meet clinically meaningful severity threshold criteria. This finding is consistent with research in childhood anxiety disorders, showing poor parent–child agreement at the diagnostic level (Comer and Kendall, 2004).

Although prior researchers have suggested that a parent-proxy report may be adequate when assessing symptoms of PTS (Clawson et al., 2013; Phipps et al., 2005), current results on categorical concordance and caseness highlight the need for allowing adolescents the opportunity to report on their own emotional experience. Failure to incorporate both parental and child/adolescent data may provide an incomplete understanding of a youth’s presenting problems, diagnosis, and appropriate treatment recommendations (Martin et al., 2004). In general, outpatient setting (both clinical and non-clinical) studies have shown only modest concordance between parents and youth on ratings of the youth’s emotional and behavioral problems (e.g. Van der Meer et al., 2008; Van Roy et al., 2010). This reduced concordance includes important intervention implications. For example, in one study, adolescents who reported clinically significant distress (while their parents did not) evidenced greater risk for mood/depressive disorders. In fact, when rating their own problems as clinically elevated, children with established mood/depressive disorders were more likely to differ from their parents’ proxy report than to be in agreement (Martin et al., 2004). When children reported greater symptoms than parents, aspects of the parent–child relationship and family structure strongly contributed to the level of concordance (Van Roy et al., 2010). Parents and children provide unique information about children’s symptoms and the impact of symptoms on functioning, suggesting that neither parent or child report can substitute for the other (Van Roy et al., 2010).

Researchers have proposed that adolescent cancer survivors would benefit from a comprehensive assessment of psychological functioning that employs multiple informants, including the adolescent, parent, and teachers (Kahalley et al., 2013). The continued distress experienced by a subset of the present sample and the moderate agreement between parent and adolescents supports this recommendation. Additionally, a recent study indicated that nurses, one of the primary conduits of referrals for patients and families, may overestimate adolescents’ ability to manage distress (Engvall et al., 2011). This finding suggests that adolescents who are experiencing significant distress may not be identified by their healthcare provider, even those with close contact, due to overestimations of emotional competence. It is imperative that the healthcare system provide outlets for adolescents to report on their emotional experience following cancer treatment.

Given our study’s small sample size, reliance on a single self-report PTSD measure, and potential for participant self-selection bias, it will be necessary to explore these findings with a larger sample and with additional measures of convergent validity. Furthermore, concordance should be evaluated with fathers and adolescents to determine whether relationships are similar. If future studies determine that parents are (relatively) under-reporting their adolescent’s symptoms of PTS distress, particularly in relation to their older adolescents, clinical implications include the need for healthcare professionals to interview adolescent cancer survivors directly about their experience of PTS symptoms and for parents to receive education about the potential for PTS symptoms in their children and how best to monitor and seek mental health services for them.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the UNM Pediatric Research Committee and the UNM General Clinical Research Center Award.

References

Achenbach

McConaughy

Howell

(1987) Child/adolescent behavioral and emotional problems: Implications of cross–informant correlations for situational specificity. Psychological Bulletin 101: 213–232.

American Psychiatric Association (2000) Diagnostic and Statistical Manual of the American Psychiatric Association (4th edn, text revision). Washington, DC: American Psychiatric Association.

American Psychiatric Association (2013) Diagnostic and Statistical Manual of the American Psychiatric Association (5th edn). Washington, DC: American Psychiatric Association.

Arpawong

Oland

Milam

et al . (2013) Post-traumatic growth among ethnically diverse sample of adolescent and young adult cancer survivors. Psycho-Oncology 22: 2235–2244.

Brown

Madan-Swain

Lambert

(2003) Posttraumatic stress symptoms in adolescent survivors of childhood cancer and their mothers. Journal of Traumatic Stress 16: 309–318.

Bruce

(2006) A systematic and conceptual review of posttraumatic stress in childhood cancer survivors and their parents. Clinical Psychology Review 26: 233–256.

Choudhury

Pimentel

Kendall

(2003) Childhood anxiety disorders: Parent-child (dis)agreement using a structured interview for the DSM-IV. Journal of the American Academy of Child and Adolescent Psychiatry 42: 957–964.

Clawson

Jurbergs

Lindall

et al . (2013) Concordance of parent proxy report and child self-report of posttraumatic stress in children with cancer and healthy children: Influence of parental posttraumatic stress. Psycho-Oncology 22: 2593–2600.

Comer

Kendall

(2004) A symptom-level examination of parent-child agreement in the diagnosis of anxious youth. Journal of American Academy of Child and Adolescent Psychiatry 43: 878–886.

10.

De Goede

IHA

Branje

SJT

Meeus

WHJ

(2009) Developmental changes in adolescents’ perceptions of relationship with their parents. Journal of Youth and Adolescence 38: 75–88.

11.

De Los Reyes

Kazdin

(2005) Informant discrepancies in the assessment of childhood psychopathology: A critical review, theoretical framework, and recommendations for further study. Psychological Bulletin 131: 483–509.

12.

Engvall

Skolin

Mattsson

et al . (2011) Are nurses and physicians able to assess which strategies adolescents recently diagnosed with cancer use to cope with disease- and treatment-related distress? Supportive Care in Cancer 19: 605–611.

13.

Erickson

Steiner

(2001) Trauma and personality correlates in long term pediatric cancer survivors. Child Psychiatry and Human Development 31: 195–213.

14.

Ghesquiere

Fan

Berliner

et al . (2008) Adolescents’ and parents’ agreement on posttraumatic stress disorder symptoms and functioning after adolescent injury. Journal of Traumatic Stress 21: 487–491.

15.

Goldstein

Wekerle

Tonmyr

et al . (2011) The relationship between post-traumatic stress symptoms and substance use among adolescents involved with child welfare: Implications for emerging adulthood. International Journal of Mental Health and Addiction 9: 507–524.

16.

Kahalley

Wilson

Tyc

et al . (2013) Are the psychological needs of adolescent survivors of pediatric cancer adequately identified and treated? Psycho-Oncology 22: 447–458.

17.

Kazak

Alderfer

Rourke

et al . (2004) Posttraumatic stress disorder (PTSD) and posttraumatic stress symptoms (PTSS) in families of adolescent childhood cancer survivors. Journal of Pediatric Psychology 29: 211–219.

18.

Landolt

Vollrath

Gnehm

et al . (2009) Post-traumatic stress impacts on quality of life in children after road traffic accidents: Prospective study. Australian and New Zealand Journal of Psychiatry 43: 746–753.

19.

Lee

Gau

Hsu

et al . (2009) A model linking uncertainty, post-traumatic stress, and health behaviors in childhood cancer survivors. Oncology Nursing Forum 36: E20–E30.

20.

Marshall

Olfson

Hellman

et al . (2001) Comorbidity, impairment, and suicidality in subthreshold PTSD. American Journal of Psychiatry 158: 1467–1473.

21.

Martin

Ford

Dyer-Friedman

et al . (2004) Patterns of agreement between parent and child ratings of emotional and behavioral problems in an outpatient clinical setting: When children endorse more problems. Journal of Developmental and Behavioral Pediatrics 25: 150–155.

22.

Meiser-Stedman

Smith

Glucksman

et al . (2007) Parent and child agreement for acute stress disorder, post-traumatic stress disorder and other psychopathology in a prospective study of children and adolescents exposed to single-event trauma. Journal of Abnormal Child Psychology 35: 191–201.

23.

Phipps

Long

Hudson

et al . (2005) Symptoms of post-traumatic stress in children with cancer and their parents: Effects of informant and time from diagnosis. Pediatric Blood and Cancer 45: 952–959.

24.

Schwartz

Drotar

(2006) Posttraumatic stress and related impairment in survivors of childhood cancer in early adulthood compared to healthy peers. Journal of Pediatric Psychology 31: 356–366.

25.

Steinberg

Brymer

Decker

et al . (2004) The University of California at Los Angeles Post-traumatic Stress Disorder Reaction Index. Current Psychiatry Reports 6: 96–100.

26.

Stuber

Kazak

Meeske

et al . (1998) Is posttraumatic stress a viable model for understanding responses to childhood cancer? Child and Adolescent Psychiatric Clinics of North America 7: 169–182.

27.

Taïeb

Moro

Baubet

et al . (2003) Posttraumatic stress symptoms after childhood cancer. European Child & Adolescent Psychiatry 12: 255–264.

28.

Van der Meer

Dixon

Rose

(2008) Parent and child agreement on reports of problem behavior obtained from a screening questionnaire, the SDQ. European Child & Adolescent Psychiatry 17: 491–497.

29.

Van Roy

Groholt

Heyerdahl

et al . (2010) Understanding discrepancies in parent-child reporting of emotional and behavioural problems: Effects of relational and socio-demographic factors. BMC Psychiatry 10: 56.