Somatic Complaints in Early Adolescence

Abstract

This study investigated the relationship between parent emotion socialization and youth somatic complaints (SC) in an early adolescent sample using a longitudinal experimental design. An emotion-focused parenting intervention, which taught parent’s skills to improve their emotional competence and emotion socialization, was used to examine whether changes in parent emotion socialization affected the young person’s SC. Questionnaires were completed prior to intervention and 10 months later by 225 youth (121 intervention) and their parents. Results indicated that changes in parents’ awareness and regulation of emotion and emotion socialization practices resulted in reduced youth SC. These findings have important implications for the prevention of SC.

Keywords

somatic complaints emotion socialization adolescence anxiety parent training

Somatic complaints (SC), such as recurrent headaches or stomach aches without any specific organic cause are common in childhood, occurring in 10% to 30% of children and adolescents (see Campo, 2012, for a review). Pain associated with SC is a source of great distress to both parents and their children and often results in considerable expense and multiple medical investigations at primary and secondary health care services (De Waal, Anrnold, Eekhof, & van Hemert, 2004; Garralda, 2011). Frequently, children’s academic and social functioning is compromised by reduced school attendance and lowered participation in extracurricular activities (Campo, 2012). To date, most studies examining familial factors in relation to child SC have compared clinical samples with “healthy” controls, and are limited by small sample sizes and populations that differ widely in age. Understanding the developmental processes involved in SC in nonclinical samples is necessary to provide guidance on the risk and protective factors that should be targeted in intervention and prevention programs (Beck, 2008).

Emotional competence, defined as the ability to recognize, understand, and regulate emotions in interpersonal and intrapersonal situations (Saarni, 1999), is thought to play an important role in the development and maintenance of SC because emotions involve the activation of physiological systems (Terwogt, Rieffe, Miers, Jellesma, & Tolland, 2006). Children’s emotional competencies are influenced by parents’ modeling of emotions, parents’ contingent reactions to children’s emotion expression, and parents’ discussions and coaching of emotions (i.e., emotion socialization practices; Morris, Silk, Steinberg, Myers, & Robinson, 2007). Parents’ deficits in emotional competence, such as lack of awareness of emotions or difficulties regulating emotions, place them at risk of modeling maladaptive emotion regulation and engaging in unsupportive or punitive socialization practices (Maliken & Katz, 2013). This, in turn, has been found to impact negatively on children’s internalizing difficulties (e.g., anxiety and depression) by inhibiting children’s adaptive affective development (Maliken & Katz, 2013; Morris et al., 2007). Youth with SC show deficits in emotional competence and higher levels of anxiety and depression (Gilleland, Suveg, Jacob, & Thomassin, 2009). Despite these findings, existing interventions involving parents, primarily aim to change parents’ responses to pain symptoms, rather than also targeting familial risk and protective factors related to children’s emotion regulation (Husain, Browne, & Chalder, 2007). Early adolescence may be an optimal time for investigating emotion socialization, as it is a time when SC are thought to peak, and when children experience an increase in the frequency and intensity of negative affect (Campo & Fritsch, 1994; Flannery, Montemayor, Eberly, & Torquati, 1993).

This study investigated whether parents’ emotion competence and their emotion socialization practices (i.e., emotion socialization) predict youth SC. Our sample was drawn from a study that investigated the efficacy of a parenting program which focused on improving emotion socialization. Therefore, we were able to research causal relations by examining whether improvements in parents’ emotion socialization led to reductions in youth SC.

Theoretical Rationale

One of the developmental tasks during childhood is to recognize that emotions have associated physical sensations, and to be able to link the physiological experience of emotions with emotion words; in turn, this aspect of emotion awareness and understanding assists with the regulation of one’s emotions (Saarni, 1999). Children and adolescents with SC have been found to have particular difficulties with identifying, differentiating and analyzing physical signals that stem from emotional arousal (Jellesma, Rieffe, Terwogt, & Kneepkens, 2006; Rieffe, Terwogt, & Bosch, 2004; Terwogt et al., 2006). They also have problems expressing negative emotions functionally (e.g., Rieffe et al., 2007), and experience negative affect more frequently and at greater intensity than children and adolescents without SC (Jellesma et al., 2006; Rieffe et al., 2004). From childhood to adulthood, higher levels of internalizing difficulties and negative affect are consistently observed in samples with clinical levels of SC when compared with healthy controls (Campo, 2012; Weersing, Rozenman, Maher-Bridge, & Campo, 2012). Difficulties with the identification and understanding of emotions, coupled with greater negative affect, are thought to undermine effective emotion regulation and lead to an increased focus on internal sensations (Berking & Wupperman, 2012; Gilleland et al., 2009).

Children learn to interpret, express, and manage their emotions largely in the context of social interactions. There is a well-established link between parenting and child/adolescent emotional competence and internalizing difficulties (see Eisenberg, Cumberland, & Spinrad, 1998; Morris et al., 2007). When parents dismiss, match, ignore, or punish emotions (i.e., emotion dismissing), children tend to have lower emotional competence and show higher levels of negative affect, anxiety, and depression. On the other hand, when parents are accepting of emotions, encourage their child’s emotion expression and respond supportively to help their child manage their emotions (i.e., emotion coaching), children tend to have better emotional competence and lower levels of anxiety or depression (Morris et al., 2007). One of the only direct investigations of the role of emotion socialization practices in child SC was conducted in India by Raval and colleagues (Raval & Martini, 2011). While Raval et al. contend that there are likely cultural differences in socialization goals (i.e., relational vs. individualistic goals) which may limit generalizability of this study to Western populations, their findings mirrored those conducted with Western populations, that have found that emotionally dismissive parenting is related to higher levels of internalizing and externalizing difficulties in children. In Raval et al.’s study, mothers of children (aged 6-8 years) with clinical levels of SC reported more negative affect and were more punitive and minimizing (i.e., emotion dismissing) of children’s negative emotion expressions, when compared with mothers of children without SC. Further, this study showed that mothers of children with clinical levels of SC reported significantly less sympathy in response to children’s anger expression compared with mothers of children who were in the clinical range for anxiety/depressive symptoms, externalizing difficulties or normal controls (Raval & Martini, 2011).

To date, research investigating familial factors in relation to children’s SC has largely focused on factors related to children’s pain behaviors rather than emotions. This research has shown that parental behaviors that may model (e.g., parents’ illness behaviors or beliefs) or reinforce pain behaviors (e.g., contingent responses, such as attention or sympathy to the child’s pain experience or allowing the child to avoid school/chores) are related to higher levels of children’s and adolescent’s SC (Jellesma, Rieffe, Terwogt, & Westenberg, 2008; Levy et al., 2004; Walker, Williams, & Smith, 2006). More consistent support has been found for the link between children’s and parents’ SC (i.e., the “modeling of illness behaviors” hypothesis) rather than parents’ contingent responses to pain. Currently, however, we know very little about why this relationship exists. The high comorbidity between SC and anxiety and depressive symptoms in children with SC suggests that additional emotion-related socializing factors may be involved. Parents with high SC may be unresponsive to children’s emotional needs, which may negatively affect the attachment relationship (Schulte & Petermann, 2011). They may engage in maladaptive emotion socialization, and dismiss children’s emotions because they are preoccupied with their own physical pain symptoms, lack emotional awareness, or find emotions difficult to manage (Berking & Wupperman, 2012; Gilleland et al., 2009; Raval & Martini, 2011). If parents dismiss children’s emotions they may inadvertently heighten children’s negative affect and anxiety and promote unhelpful emotion-regulation strategies (e.g., suppression or avoidance), which may, in turn, increase the likelihood that children experience prolonged negative affect, placing them at risk of experiencing SC. In addition, if children gain greater parental attention when talking about their physical symptoms (rather than their emotions), they may not learn to identify their internal states accurately. Interestingly, compared with healthy controls, children with clinical levels of SC have been found more likely to endorse beliefs that emotions are insignificant or trivial and communicating them is useless (Raval, Martini, & Raval, 2010).

Other familial risk factors related to children’s SC are parental psychopathology, dysfunctional family climate, and insecure attachment (Schulte & Petermann, 2011), all of which have been found to be related to parents’ deficits in emotional competence and maladaptive parenting and may be able to be targeted via group parenting programs (Maliken & Katz, 2013). Yet, there are currently no published evaluations pertaining to prevention or treatment of child/youth SC that target familial risk factors via group-based parent training. This is surprising, given the prevalence of SC is similar to that of other internalizing difficulties and parental factors play an important role in the development and maintenance of SC (Campo, 2012; Schulte & Petermann, 2011). In addition, treatment of children’s SC has primarily focused on teaching self-management of pain rather than on managing emotions (Husain et al., 2007). Thus, when parents are involved in their child’s treatment, parental components of interventions often teach parents to ignore mild pain symptoms and to encourage the child’s engagement in routine activities. Focusing solely on pain symptoms and not also teaching parents how to recognize and validate the child’s emotions (aspects of adaptive emotion socialization) may lead to missed opportunities for parents to foster the young persons’ healthy affective development.

One parenting intervention that has been found to be efficacious in improving parents’ emotion socialization is the Tuning in to Kids program (TIK; Havighurst, Wilson, Harley, Prior, & Kehoe, 2010). A variant of this program, called Tuning in to Teens (TINT; Havighurst, Harley, Kehoe, & Pizarro, 2012), has been adapted for parents of adolescents. Both TIK and TINT aim to increase children and young people’s emotional competencies via improving parents’ emotion socialization (see Havighurst, Wilson, Harley, & Prior, 2009; Kehoe, Havighurst, & Harley, 2014, for a description of the program including theoretical underpinnings). Parents are taught skills to improve their own emotional awareness and regulation, increase empathic responding, and reduce dismissive responding to children’s emotions (i.e., an emotion coaching style; Gottman, Katz, & Hooven, 1997). An evaluation of the impact of TINT on youth internalizing problems (using the same sample) showed reductions in parents’ difficulties with emotion awareness and regulation, emotion dismissing parenting responses and young people’s internalizing problems (anxiety and depressive symptoms) 10 months later (Kehoe et al., 2014). The current study examined whether these changes in parents’ emotion socialization would also lead to changes in youth SC. This would allow examination of emotion socialization as a causal factor in youth SC.

Aims of the Current Study

Research investigating the relationship between parent emotion socialization and youth SC is in its infancy. Therefore, to extend research by Raval et al. (Raval & Martini, 2011), using a Western population, and controlling for youth negative affect and anxiety, we first used baseline data to investigate our research question:

Research Question 1: Do parental SC, parent difficulties with emotion awareness and regulation and parents’ emotion dismissive responses uniquely explain variance in youth SC over and above the variance explained by youth negative affect and anxiety?

Second, to allow examination of emotion socialization as a causal factor of youth SC, we investigated the efficacy of TINT in reducing youth SC using both baseline and 10 months follow-up data. We hypothesized that there would be greater reductions in youth SC for participants in the intervention condition compared with the control condition. Finally, to explain causal relations between emotion socialization and youth SC, we investigated whether improvements in youth SC could be explained by changes in parents’ emotion socialization. Specifically, we hypothesized that reductions in parents’ difficulties with awareness and regulation of emotion and emotion dismissing parenting would lead to reductions in youth SC.

Method

Participants

Participants were Grade 6 students aged 10 to 13 years ( $\bar{X}$ = 12.01, SD = 0.42) and their primary caregiver (200 mothers, 25 fathers; $\bar{X}$ _age = 44.1, SD = 5.13). Fathers selected themselves into the study as “primary caregivers” and did not differ from mothers on any of the variables apart from rating themselves as more emotionally dismissive ( $\bar{X}$ = 2.28, SD = 0.30) compared with mothers ( $\bar{X}$ = 2.02, SD = 0.38); t(219) = 4.01, p < .001, 95% CI = [0.13, 0.39]. Fathers were retained in the study because (a) fathers were primary caregivers, (b) fathers in each condition did not differ in levels of emotion dismissing, and (c) fathers have generally been found to be more dismissing of emotion with their children, but fathers’ responses to their adolescents’ emotions have been found to be important for healthy development of emotional competence (Brand & Klimes-Dougan, 2010). One parent withdrew consent for their child to participate in the study prior to questionnaire completion, leaving 225 parent and 224 youth participants (48.7% boys). Parents were predominantly Caucasian (89.8%) and Asian (8%), mostly married/defacto (82.7%), 28 parents (12.4%) had repartnered, 5 (2.2%) were sole parents, and 5 (2.7%) had not reported their marital status. Most parents spoke English as their first language (81.3%) and had completed high school (75.6%); 38.2% had completed a non-university qualification, and 43.5% had completed a bachelor’s degree or higher. Parents were mostly in paid employment (80.4%), worked 29.5 hour per week on average (SD = 11.0), and reported gross annual combined family incomes of less than AUS$40,000 (9.8%), AUS$40,000 to AUS$59,999 (15.1%), AUS$60,000 to AUS$99,999 (32.0) and AUS$100,000 or more (38.7%). Ten parents did not report their income (4.4%). Participants’ education, employment, and income levels were consistent with the normal distribution for Australia (Australian Bureau of Statistics, 2011). Youths with a score of two, indicating a definite presence of SC (Steinhausen & Winkler Metzke, 2007) on nine SC items (Achenbach, 1991) were as follows: dizzy (9%), overtired (14.7%), aches and pains (7.6%), headaches (6.2%), nausea (4%), eye problems (6.2%), skin problems (6.2%), stomach ache (3.6%), vomiting (0.4%). Youth anxiety was also representative of the normal population, with 8.9% of the young people reporting levels in the clinical range and mean parent-reported youth anxiety comparable with normal population means (Nauta et al., 2004; Spence, Barrett, & Turner, 2003).

Procedure

The study was conducted from 2008 to 2010 (see Kehoe et al., 2014, for a detailed description). Briefly, using a computer randomizer, schools (n = 120) were randomized (ratio of 1:1) into either intervention (TINT Parenting program) or control condition (questionnaire study on transition to secondary school). Group randomization minimizes contamination and is preferable when investigating the effects of an intervention delivered to identifiable groups (Murray, 1998). After randomization, schools were approached, and 55 schools (45%; 28 intervention, 27 control) agreed to participate. Reasons for nonparticipation of schools included a predominance of non-English speaking families at the school, and commitment to other research projects and parenting/transition programs. A total of 3,359 information letters and consent forms (1,759 intervention) attached to the school newsletter were sent home to Year 6 students and their parents. Letters were identical for each condition (explaining that we were researching factors related to successful transition to secondary school); however, intervention parents were also invited to participate in a six-session parenting program. Three hundred twenty-three parent–student dyads (171 interventions) returned signed consent forms and were sent a survey package, translating to a 9.25% response rate. This represents a comparable initial response rate with other universal parenting program recruitment efforts for parents of adolescents which typically recruit in schools (e.g., Ralph & Sanders, 2006). Incentives were offered to participating youth in the form of a AUS$30 CD voucher receivable on completion of follow-up questionnaires. Parents and youth were asked to complete questionnaires separately and provided with individual envelopes to ensure their privacy. A total of 229 parent–youth dyads (70.9%, 125 interventions) returned completed baseline questionnaires and were assessed for eligibility. Four parents allocated to the intervention did not proceed (group did not go ahead due to insufficient number of parents, n = 3, and times of program did not suit parent, n = 1). These parents were provided with alternative support options and were not included in the study. Follow-up was conducted on average 10.5 months (SD = 0.74, range 10-12 months) post baseline when youth were at secondary school (seventh grade). The study conformed to all ethical requirements for research.

A total of 12 TINT parenting groups (group size between 6 and 13 parents) were conducted during four terms (autumn, winter, spring, summer) in the first 2 years of the study. The program was delivered in the evening for 6 consecutive weeks (2 hour per week). Two facilitators (one of whom was the first or the third author) used a structured manual (Havighurst et al., 2012) delivering 100% of the compulsory content for all groups. High attendance was recorded, with 84% of parents completing four or more sessions (31% of parents attended all six sessions). Four parents only attended one session due to work commitments (n = 2), child care issues (n = 1), and a family crisis (n = 1). If parents missed sessions, they were sent materials and phoned to discuss any questions or concerns. The study had excellent retention, with only 15 of intervention and control participants failing to return questionnaires at follow-up. These participants did not significantly differ from the rest of the sample on any of the measures and there was no significant difference in questionnaire return rate between the intervention (n = 111, 91.7%) and control group dyads (n = 99, 95.2%), χ²(1, n = 225) = 0.59, p = .442, phi = −.07.

Measures

Youth SC (parent and youth report)

Frequency of youth SC were measured at baseline and 10 months post intervention using the 9-item SC subscale (rated 0 = not true to 2 = very true or often true) from the Child Behavior Checklist (CBCL; parent report) and the Youth Self-Report version (Achenbach, 1991). Internal consistency, test–retest reliability and validity of the CBCL is well established (Achenbach & Rescorla, 2001) and was comparable in the current study (Cronbach’s α values were .69 and .70 at baseline and .67 and .76 at follow-up for the parent and youth versions, respectively). Interrater agreement (parent–youth) was moderate (see Table 1).

Table 1.

Means, Standard Deviations, and Bivariate Correlations for Parent and Youth Variables at Baseline.

	Subscale	$\bar{X}$ (SD)	1	2	3	4	5	6	7
1	Youth SC (Y)	4.25 (3.0)	—
2	Youth SC (P)	2.71 (2.3)	.37***	—			—	—	—
3	Youth negative affect (Y)	6.8 (2.3)	.56***	.31***	—		—	—	—
4	Youth anxiety^a (Y)	22.19 (13.0)	.55***	.32***	.63***	—
5	Parental SC (P)	4.95 (3.5)	.06	.25***	.18**	.07	—	—	—
6	Parents’ difficulties AW/ER^a (P)	69.81 (20.1)	.11	.32***	.21**	.17*	.41***	—	—
7	Emotion dismissing (Y)	2.17 (0.3)	.23***	.11	.28***	.34***	.03	.19**	—
8	Emotion dismissing (P)	2.05 (0.4)	.04	.09	.14*	.15*	.09	.38***	.40***

Note. SC = somatic complaints; (Y) = youth-rated; (P) = parent-rated; AW/ER = emotion awareness and regulation.

Transformed variable.

p < .05. **p < .01. ***p < .001.

Youth anxiety (youth self-report)

Baseline youth anxiety symptoms were measured with the Spence Children’s Anxiety Scale (SCAS; Spence, 1998). The SCAS is widely used and has shown excellent reliability (including test–retest reliability) and validity (Nauta et al., 2004). In the current study, the total scale was used to reduce multiple comparisons. Cronbach’s alpha coefficient was .90 at baseline indicating high internal consistency.

Youth negative affect (youth self-report)

Baseline frequency of negative affect was measured with the three global items from the Emotions as a Child Scales (EAC; O’Neal & Magai, 2005). The three items ask young people for each emotion (sadness, anger, anxiety): “Over the past month how often did you feel . . . (sad or down/angry or frustrated/fearful or anxious)?” Items were scored individually on a 5-point Likert-type scale (1 = never to 5 = very often) and summed to create a total negative affect score. Larger scores indicate higher frequency of negative emotions. Cronbach’s alpha was .78, and corrected item intercorrelations were between .58 and .67 indicating good reliability for a scale with three items (Pallant, 2007).

Parental SC (parent report)

Baseline levels of parental SC were measured using the SC subscale (7 items, rated 0-3) from the General Health Questionnaire (GHQ 28; Goldberg, 1981), with higher scores indicating higher presence of parental SC. The GHQ has been widely used as a general screening measure of psychological well-being and has shown good internal consistency and test–retest reliability in previous studies (Andersen, Sestoft, Lillebaek, Gabrielsen, & Hemmingsen, 2002). Cronbach’s alpha for the SC subscale was .82.

Parents’ difficulties in emotion awareness and regulation (parent report)

The Difficulties in Emotion Regulation Scale (DERS; Gratz & Roemer, 2004) was used to assess parents’ own difficulties with emotion awareness/regulation. The scale is a 36-item self-report questionnaire rated from 1 (almost never) to 5 (almost always) and measures difficulties in: acceptance of emotions, ability to engage in goal-directed behavior when distressed, impulse control, awareness of emotions, access to strategies for regulation, and clarity of emotions. Higher scores indicate greater difficulties in emotion awareness/regulation. The DERS has demonstrated very good overall internal consistency and test–retest reliability. In the current study, Cronbach’s alpha for the total scale was .94 (baseline) and .93 (follow-up).

Parents’ emotion socialization strategies (parent and youth report)

The 45-item EAC (Magai, 1996; Magai & O’Neal, 1997) was used to measure parent emotion socialization strategies in response to the young person’s negative emotion expressions (anger, sadness, fear). Parent- and youth-report forms are identical, varying only according to the phrasing of questions from the parent or the young person’s perspective. The EAC measures five emotion socialization responses (a total of nine items per subscale, which includes three items for each emotion, ranging from 1 = never to 5 = very often): for example, “When my child was angry/sad/fearful . . . ” “ . . . I asked my child what made her/him mad/sad/fearful” (encouragement response); “ . . . I told my child that s/he was acting younger than his/her age” (punishing response); “ . . . I did not pay attention to her/his being angry/sad/fearful” (neglect response); “ . . . I got very angry/sad/fearful” (matching/magnifying response); and “ . . . I told her/him to cheer up” (overriding response). Acceptable internal consistency, test–retest reliability, and validity statistics for various versions of the EAC with both adolescent and adult samples have been found (Garside & Klimes-Dougan, 2002; O’Neal & Magai, 2005; Silk et al., 2011). To reduce multiple comparisons and remedy problems of multicollinearity (several subscales correlated > .60), the total emotion dismissing response variable was used. For the total scale, the nine encouragement of emotions items are reverse scored, and items for all five subscales are combined and averaged. High scores reflect greater dismissing of anger, sadness and anxiety. In the present study, Cronbach’s alpha coefficients for the total emotion dismissing scale were .87 and .88 (baseline) and .80 and .85 (follow-up) for the parent and youth versions, respectively.

Results

Analytic Strategy

Preliminary analyses included bivariate correlations and hierarchical regression analyses to investigate the relationship between the variables at baseline; and independent samples t tests and chi-square tests to examine condition comparability and determine covariates to be included when examining the efficacy of the intervention (Pocock, Assmann, Enos, & Kasten, 2002). In all analyses, we controlled for youth negative affect and anxiety to allow examination of unique contributions of parent variables to youth SC. Due to the multistage sampling strategy, where schools were sampled first, followed by the parent–youth dyads within these schools, intraclass correlations (ICC) were computed. The computation of ICC enables the variance in the baseline variables affected by school membership to be assessed. Results showed ICC ranged between .000 and .015, suggesting that higher level grouping (by school) did not affect the estimates in any meaningful way, and single-level analyses were appropriate for baseline analyses (Heck, Thomas, & Tabata, 2010). However, for examination of intervention effects, multilevel linear mixed model regression analyses were conducted (SPSS version 18) to examine the effects of treatment condition over time on youth SC, while controlling for school-related influence on intervention outcomes (Heck et al., 2010). Best model fit was achieved using maximum likelihood and a variance components covariance structure, with intercept and school as random effect and time and covariates as fixed (Heck et al., 2010). Thus, the dependent variable was regressed on time, treatment condition and their interaction while controlling for covariates. Interpretation of the results focused on results for the Time × Condition interaction (rather than main effects) to examine whether greater change was found for intervention as compared with the control participants.

Effect sizes were calculated using the recommended formula for multilevel analysis for controlled clinical trials (Feingold, 2009). Effect sizes (d) greater than .8 are generally considered large, while those equal to .5 are moderate and .2 are small (Cohen, 1988).

To investigate whether the TNT program led to changes in youth SC via changes in parents’ emotion socialization, a series of mediation analyses were conducted. Because mediation analyses produced identical results for the parent and youth model, composite scores were created with unit weighted z scores (e.g., composite emotion dismissing = z_{youth-reported emotion dismissing} + z_{parent-reported emotion dismissing}; Warner, 2012), and the results presented are for combined scores. As recommended by Preacher and Hayes (Preacher & Hayes, 2008), indirect effects were calculated by bootstrapping (with 1,000 resamples). With this approach, confidence intervals are computed by repeatedly sampling from the data set and estimating the indirect effect in each set, adjusting for bias and skewness in the bootstrap distribution; the statistical significance of the a (independent variable to mediator) and b (mediator to the dependent variable) paths is not considered. Instead, the interpretation of the mediation analysis focused on the 95% confidence interval, where the absence of zero in the interval indicated that the indirect effect was statistically significant. Given that the independent variable (condition) was not experimentally manipulated, we predicted follow-up outcomes, using follow-up parenting variables as mediators, while controlling for baseline outcome and mediator variables (A. F. Hayes, personal communication, September 19, 2012). For example, when assessing whether parents’ emotion dismissing mediated program effects on youth SC at 10 months follow-up, “Condition” was used as the independent variable, and follow-up scores on parents’ emotion dismissing was used as the mediator, while controlling for baseline youth SC and baseline parents’ emotion dismissing.

Preliminary Analyses

Due to violation of normality, square root and log transformations were applied to the variables parental SC (skewness: from 1.46 to 0.72, and kurtosis: from 2.29 to 0.43) and parent emotion awareness/regulation (skewness: from 1.12 to 0.17; kurtosis: from 2.99 to 0.24). For ease of interpretation, untransformed means are presented, but statistics given are for analyses conducted with transformed variables.

Bivariate correlations using baseline data showed different patterns of relations among the sets of predictors and dependent variables, depending on whether youth or parents were reporting (see Table 1). Youth-reported SC correlated positively with parent-reported youth SC, negative affect, anxiety, and parent emotion dismissing. Parent-reported youth SC correlated positively with negative affect, anxiety, parental SC, and parents’ difficulties with emotion awareness/regulation. Contrary to expectations, youth-reported SC were unrelated to parents’ difficulties with emotion awareness/regulation and parental SC, and parent-reported youth SC were unrelated to parent-reported parent emotion dismissing. Parent emotion dismissing and parents’ difficulties in emotion awareness/regulation were, however, positively correlated with youth negative affect and anxiety (parent and youth report). None of the demographic variables correlated with parent or youth variables.

Next, hierarchical regression analyses were conducted which examined if parent emotion dismissing (youth model) or parental SC and parents’ difficulties in emotion awareness and regulation (parent model) predicted unique variance in youth SC. To control for youth-reported anxiety and negative affect, these variables were entered at Step 1, followed by parent emotion dismissing (youth model) or parental SC and parent difficulties in awareness and regulation (parent model) at Step 2. The youth-report model was significant at Step 1, F(2, 218) = 65.77, p < .001; and Step 2, F(3, 218) = 43.78, p < .001, and the final model explained 38% of the variance in youth SC. However, only youth negative affect and anxiety were significant predictors of youth SC (see Table 2). In the parent-report model, parental SC, parents’ difficulties with emotion awareness/regulation significantly improved the model when entered at Step 2. The model was significant, F(4, 221) = 13.77, p < .001, and explained 20% of the variance in youth SC. Only parent difficulties with emotion awareness/regulation and youth anxiety remained significant predictors of youth SC (see Table 2).

Table 2.

Hierarchical Multiple Regression Predicting Baseline Youth Somatic Complaints.

	Predictor	b	SE b	β	R ²	ΔR²
Youth somatic complaints (Y)
Step 1
	Youth negative affect	0.45	0.09	.35***
	Youth anxiety^a	0.71	0.15	.33***	.38
Step 2
	Youth negative affect	0.44	0.09	.35***
	Youth anxiety^a	0.69	0.15	.32***
	Parent emotion dismissing (Y)	0.26	0.51	.23	.38	.001
Youth somatic complaints (P)
Step 1
	Youth negative affect	0.19	0.08	.19*
	Youth anxiety^a	0.32	0.14	.19*	.12	—
Step 2
	Youth negative affect	0.13	0.08	.12
	Youth anxiety^a	0.32	0.13	.19*
	Parental somatic complaints^a	0.54	0.23	.15*
	Parent difficulties AW and ER^a	4.17	1.40	.20**	.20	.09

Note. All variables measured at baseline. (Y) = youth-rated; (P) = parent-rated; AW and ER= emotion awareness and regulation.

Transformed.

p < .05. **p < .01. ***p < .001.

Intervention Analyses

Next, we investigated whether the TINT parenting program was efficacious in reducing youth SC. Between-group comparisons of baseline data found no statistically significant differences in the scores for any of the youth-reported variables under study. However, parent data showed, at baseline, that intervention parents rated their youth significantly higher on SC, t(221) = 3.50, p = .001, compared with control parents. In addition, youth in the intervention group were slightly younger, t(223) = −6.24, p < .001 (mean difference in months = −.36), and there were more boys (n = 67, 55.4%) in the intervention group, compared with the control group (n = 42, 40.4%), χ²(1, n = 225) = 4.48, p = .035, phi = .15. Youth age differences may have occurred due to significantly more baseline data being collected for control participants in the second half (Terms 2, 3, and 4) of the Grade 6 school year, whereas data were collected throughout the year (Terms 1-4) for the intervention group, χ²(3, n = 225) = 57.34, p < .001, phi = .51. In addition, participants enrolled in the study in Term 1 (n = 31) were followed up slightly later ( $\bar{X}$ _months = 12.00, SD = 0.00) compared with participants who enrolled in the study in Terms 2, 3, or 4 ( $\bar{X}$ _months = 10.3, SD = 0.46), t(179) = 50.06, p < .001, 95% CI = [1.6, 1.8]. Hence, parent-rated youth SC, gender, the young person’s age, and baseline term of enrollment were all included as additional covariates. Gender and baseline term of enrollment were also included as moderators, but interactions were not significant and are not reported.

Multilevel mixed effects modeling

For youth-reported youth SC, the main effect of time and the interaction between time and condition were not significant, indicating that young people from both conditions reported no change in SC. For parent-reported youth SC, a significant Time × Condition interaction was found (β = −.66, SE = .29, df = 208.9, t = −2.25, 95% CI = [−1.23, −0.08], p = .025, d = .28), indicating greater reductions for the intervention group from baseline ( $\bar{X}$ = 3.32, SE = .21) to follow-up ( $\bar{X}$ = 2.70, SE = .21), compared with control parents who reported no change from baseline ( $\bar{X}$ = 1.95, SE = .23) to follow-up ( $\bar{X}$ = 1.98, SE = .23).

Finally, examination of mediators of treatment effects showed both parents’ difficulties in emotion awareness and regulation and parents emotion dismissing were found to mediate the effect of condition on youth SC at follow-up (see Table 3).

Table 3.

Mediation of the Effect of the Intervention on Youth Somatic Complaints through Parent Emotion Socialization Variables.

Outcome	Mediator	Point estimate	SE	BCa [95% CI]
Youth SC (C)	Parental AW and ER^a	−.12*	.07	[−0.234, −0.017]
Youth SC (C)	Parents’ emotion dismissing (C)	−.30*	.10	[−0.467, −0.154]

Note. BCa = bias corrected and accelerated; SC = somatic complaints; C = parent-rated and youth-rated composite score; AW and ER = parent emotion awareness and regulation.

Transformed variable.

Significant indirect effect.

Discussion

This study explored the relationship between parent emotion socialization and youth SC. To date, most studies examining familial factors in relation to child or adolescent SC have compared clinical samples with “healthy” controls using cross-sectional designs. In contrast, the current study was conducted with a large community sample, using an experimental longitudinal design, with both parent- and youth-reported data. Further, this study was the first to investigate whether a group parenting program, which aims to improve emotion socialization, would reduce youth SC. The hypothesis that greater reductions in youth SC would be found for participants in the intervention condition as compared with the control condition was partially supported. In addition, the hypothesis that reductions in parents’ awareness and regulation of emotion and emotion dismissing parenting would lead to reductions in youth SC was supported. The study extends the literature on SC by considering parents’ emotional competence and emotion socialization practices as predictors of youth SC alongside parents’ own SC.

To examine whether parenting variables explained unique variance over and above youth negative affect and anxiety, bivariate correlations and two hierarchical regression analyses were conducted using the baseline data. Relations between variables differed depending on who reported: youth or parents. In the youth-reported model, bivariate correlations revealed significant correlations between parents’ emotion dismissing and youth SC. Regression analyses showed, however, when controlling for negative affect and anxiety, parent’s emotion dismissing was no longer a significant predictor. Youth negative affect and anxiety have been identified as important risk factors for young people’s SC (Campo, 2012; Gilleland et al., 2009; Weersing et al., 2012) and in the current study were the strongest predictors of youth-reported SC. The results suggest that emotion socialization may exert its influence on youth SC indirectly, via youth negative affect and anxiety. When parents use more emotionally dismissive parenting this may contribute to greater negative affect and anxiety in the young person, which may lead to more somatic difficulties. This suggestion is supported by research that has found parent emotion dismissing practices are related to higher levels of young people’s negative affect and anxiety (Morris et al., 2007) and youth with higher levels of internalizing difficulties and negative affect report higher SC (Campo, 2012; Weersing et al., 2012). Raval and colleagues also found in India, that parents of children with clinical levels of SC were less sympathetic and more emotionally dismissive in response to children’s negative emotion when compared with normal controls (Raval & Martini, 2011). Our study provides support for Raval et al.’s findings using a Western population. The findings highlight the importance of considering and measuring youth negative affect and anxiety when investigating or intervening with youth SC. They also support the targeting of parental factors that are related to youth anxiety and negative affect when treating or preventing SC.

For the parent-report model, parents’ difficulties with emotion awareness and regulation, parental SC and youth anxiety remained significant predictors of youth SC. Parents’ difficulties with emotion awareness and regulation emerged as the most important predictor of parent-reported youth SC, over and above parental SC and youth negative affect and anxiety. Interestingly, despite the link between deficits in emotional competence and SC in adult populations (Campo, 2012), no other study has examined parents’ difficulties in aspects of emotional competence as a predictor of children’s SC. Consistent with prior research, bivariate correlations showed parents’ difficulties in emotion awareness and regulation were related to greater parental SC (Campo, 2012) and greater parent- and youth-reported emotion dismissing responses (Morris et al., 2007). When parents are unable to manage their own emotions effectively they are likely to have difficulties accessing strategies to work through emotional problems, including with their adolescents. This may result in youth not learning adaptive problem solving around emotional situations and this, in turn, may affect their emotion understanding and regulation resulting in heightened negative affect (Maliken & Katz, 2013). In addition, parents’ modeling of dysregulation may reflect parental psychopathology, or contribute to a dysfunctional family climate and insecure attachment, each of which has been found to be a risk factor for youth SC (Maliken & Katz, 2013; Schulte & Petermann, 2011). Although not examined in this study, parents’ difficulties in awareness and regulation are also likely to influence parents’ contingent responses to children’s pain experience, which has been found related to child SC (e.g., Levy et al., 2004). If children’s complaints about pain arouse negative emotions in the parent, a parent who finds emotions difficult to tolerate may choose a solution that will reduce negative affect more immediately (i.e., allow the child to avoid school or chores). Our findings highlight the importance of considering parents’ emotional competence when examining familial factors related to youth SC.

In the extant literature, there has been a significant focus on the relationship between parental SC and child/youth SC, because parents’ SC are thought to effect their children’s attitudes and beliefs about illness. Similar to other studies, the current study found parental SC predicted parent-reported youth SC (Gilleland et al., 2009; Levy et al., 2004). In contrast to Levy et al. (2004), but similar to Gilleland et al. (2009), we did not find a relationship between parental SC and youth-reported youth SC. When parents’ experience greater levels of SC they may be more self-focused and therefore may not notice the young person’s emotions or they may be more reactive in response to the young person’s negative emotions due to being in pain or not having energy to respond. In turn, young people may feel overlooked, frustrated, or sad about this lack of support leading to higher levels of negative affect. We found that higher rates of parental SC were related to higher levels of youth negative affect (which included sadness and anger), which supports this hypothesis. Finally, it is possible that parents with higher levels of SC themselves are more empathic with the young persons’ pain experience. Future studies should investigate if parents of children with SC are indeed more responsive when children express their emotions by describing the physical sensations versus when they express emotions using emotion labels.

Examination of the impact of the TINT program on youth SC allowed us to study whether changes in parent emotion socialization would lead to changes in youth SC. A significant interaction between time and condition was found for parent-reported youth SC, indicating greater change for those in the intervention condition, who reported significantly lower youth SC at follow-up compared with control parents who reported no change. The effect size was small to medium. This finding highlights that an intervention which targets parent emotion socialization can lead to a reduction in parent-reported youth SC. Although young people’s report of their SC did not change, baseline analyses found that youth anxiety and negative affect were important predictors of youth SC. In other research, we have found a reduction in youth-reported anxiety after parents completed the emotion socialization program (Kehoe et al., 2014). Perhaps, greater change happens in youth perceptions of their anxiety which may over time contribute to changes in SC, in particular to changes in duration or intensity of SC and subsequent impairment. It is also possible that the frequency, duration or intensity of the pain may have been reduced at home (due to parental emotion coaching), but not at school. We did not, however, assess intensity or duration of SC, nor did we assess functional impairment or visits to pediatric clinics or contact with school nurses during school hours. Hence, it is not known whether youth-reported SC occurred at home or at school, or whether there were changes in duration or intensity of SC. Further research should investigate if the intervention reduces SC using these other measures. Another possibility is that, after taking part in the intervention, parents in the current study may have been better able to recognize emotions in their children and attribute emotional distress as a cause of the pain, therefore reporting fewer symptoms “of unknown origin.” Nevertheless, youth anxiety and negative affect were important predictors of youth SC, and changes in parents’ emotion socialization resulted in reduced parent-reported youth SC. Therefore, both (a) parents ability to respond supportively to negative emotion expression and (b) the young persons’ ability to recognize, differentiate, understand, express, and manage their emotions may prove fruitful additions to existing intervention or prevention programs that target child or youth SC, and warrant further investigation.

Limitations of the Current Study

A number of limitations need to be considered when interpreting these findings. First, to assess youth SC, a parent and youth-reported questionnaire was used, and so intervention outcomes may have been affected by an expectancy bias, especially on parent report. Observation of parent emotion socialization and independent medical evaluations to verify changes in SC would have strengthened the study. Second, sampling of frequency, intensity and duration of SC would have strengthened measurement and provided clearer information on which dimensions of SC were most impacted by changes in parents’ emotion socialization. Third, although the randomization resulted in two groups who were comparable across socioeconomic status and youth-reported variables, participants were not blind to condition and parents who selected to take part in the intervention reported greater youth SC at baseline. Fourth, mediation analyses were conducted using a “half-longitudinal design,” which, although superior to meditational analyses with cross-sectional data, is inferior to meditational analyses that use a full longitudinal design (Cole & Maxwell, 2003). This is because, unless there are three time points and the mediator is directly manipulated, temporal order of the mediator to outcome pathway cannot be established with certainty. However, even with three time points, other potential variables (e.g., improved youth emotion regulation, attachment relationships, or family conflict) may mediate the effect of the program (Bullock, Green, & Ha, 2010; Cole & Maxwell, 2003). A strength of the current study was that only the proposed mediators were a direct target of the intervention (Bullock et al., 2010). Therefore, it is likely that the proposed direction of effects is the most plausible explanation of the mechanisms of change. Fifth, the sample contained few families from multicultural or very low socioeconomic status backgrounds which may limit generalizability of the findings. However, given that the sample was comparable with the general population of Australia with regard to socioeconomic status variables (Australian Bureau of Statistics, 2011), youth levels of anxiety (Nauta et al., 2004), and SC (Steinhausen & Winkler Metzke, 2007), and that up-take rates were comparable with other universal parenting prevention programs (Ralph & Sanders, 2006), it is likely that the sample represents families who may typically participate in community based research and preventive interventions.

Conclusion

Children and adults with SC show higher levels of internalizing difficulties and negative affect when compared with healthy controls, which, in part, has been attributed to difficulties in emotional competence. This study adds to research that has linked emotional competence and SC by investigating the role of parental emotion socialization. The results indicated that changes in parents’ awareness and regulation of emotion and emotion socialization practices led to changes in youth SC, when controlling for youth anxiety and negative affect. This study found support for the use of an emotion-focused group parenting intervention, which targets parents’ emotion socialization for prevention of youth SC.

Footnotes

Acknowledgements

We thank the staff from ParentsLink as well as Lara Silkoff, Dr. Carol Hulbert, Dr. Paul Dudgeon, and Dr. Andrew Hayes. We extend our thanks to all the schools, parents, and children who participated in the study. We would also like to thank anonymous reviewers for their helpful comments.

Authors’ Note

This study has been a partnership between The University of Melbourne and ParentsLink at MacKillop Family Services.

Declaration of Conflicting Interests

The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: The authors wish to disclose a conflict of interest in that they might profit from positive reported outcomes of evaluation of the Tuning in to Teens parenting program.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The authors received financial support for the research from Mindful, Centre for Training and Research in Developmental Health, Department Psychiatry, The University of Melbourne. Parentslink, at McKillop Family Services provided financial support relating to food and venue hire for parent groups.

Author Biographies

Christiane E. Kehoe, BSSc (Hons), is a PhD candidate at the University of Melbourne, Department of Psychology and coauthor of the Tuning in to Teens parenting program. She has been working at Mindful, Department of Psychiatry, University of Melbourne, as part of the Tuning in to Kids/Teens research and training team and works privately as a parent and teacher consultant. She is particularly interested in the prevention of internalizing and externalizing problems in children and adolescents.

Sophie S. Havighurst works as a senior lecturer and clinical psychologist at Mindful, Department of Psychiatry, University of Melbourne. For many years, she has worked on the development and evaluation of the Tuning in to Kids parenting program, which involves the application of theories about parent emotion socialization to interventions that promote children’s emotional competence, social and behavioral functioning. She is particularly interested in the field of prevention and early intervention.

Ann E. Harley, MEd, Grad Dip Ed, Dip Teach, BA, is the training manager and coauthor of the Tuning in to Kids and Tuning in to Teens parenting programs. She has more than 30 years of experience in parent education and counseling, including working with families from lower socioeconomic and culturally and linguistically diverse communities. Along with training professionals and delivering parenting programs, she also provides supervision and specialist consultation to trainees and agencies providing parenting services.

References

Achenbach

T. M.

(1991). Manual for the Child Behavior Checklist/4-18 and 1991 profile. Burlington: Department of Psychiatry, University of Vermont.

Achenbach

T. M.

Rescorla

L. A.

(2001). Manual for the ASEBA school-age forms & profiles. Burlington: Department of Psychiatry, University of Vermont.

Andersen

H. S.

Sestoft

Lillebaek

Gabrielsen

Hemmingsen

(2002). Validity of the General Health Questionnaire (GHQ-28) in a prison population: Data from a randomized sample of prisoners on remand. International Journal of Law and Psychiatry, 25, 573-580. doi:10.1016/S0160-2527(01)00085-1

Australian Bureau of Statistics. (2011). 2009-10 household income and income distribution. Canberra: Commonwealth of Australia.

Beck

J. E.

(2008). A developmental perspective on functional somatic symptoms. Journal of Pediatric Psychology, 33, 547-562. doi:10.1093/jpepsy/jsm113

Berking

Wupperman

(2012). Emotion regulation and mental health: Recent findings, current challenges, and future directions. Current Opinion in Psychiatry, 25, 128-134. doi:1097/YCO.0b013e3283503669

Brand

A. E.

Klimes-Dougan

(2010). Emotion socialization in adolescence: The roles of mothers and fathers. New Directions for Child and Adolescent Development, 128, 85-100. doi:10.1002/cd.270

Bullock

J. G.

Green

D. P.

S. E.

(2010). Yes, but what’s the mechanism? (Don’t expect an easy answer). Journal of Personality and Social Psychology, 98, 550-558. doi:10.1037/a0018933

Campo

J. V.

(2012). Annual research review: Functional somatic symptoms and associated anxiety and depression—Developmental psychopathology in pediatric practice. Journal of Child Psychology and Psychiatry, 53, 575-592. doi:10.1111/j.1469-7610.2012.02535.x

10.

Campo

J. V.

Fritsch

S. L.

(1994). Somatization in children and adolescents. Journal of the American Academy of Child & Adolescent Psychiatry, 33, 1223-1236. doi:10.1097/00004583-199411000-00003

11.

Cohen

(1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Lawrence Erlbaum.

12.

Cole

D. A.

Maxwell

S. E.

(2003). Testing mediational models with longitudinal data: Questions and tips in the use of structural equation modeling. Journal of Abnormal Psychology, 112, 558-577. doi:10.1037/0021-843X.112.4.558

13.

De Waal

M. W.

Anrnold

I. A.

Eekhof

J. A.

van Hemert

A. M.

(2004). Somatoform disorders in general practice: Prevalence, functional impairment and comorbidity with anxiety and depressive disorders. British Journal of Psychiatry, 184, 470-476. doi:10.1192/bjp.184.6.470

14.

Eisenberg

Cumberland

Spinrad

T. L.

(1998). Parental socialization of emotion. Psychological Inquiry, 9, 241-273. doi:10.1207/s15327965pli0904_1

15.

Feingold

(2009). Effect sizes for growth-modeling analysis for controlled clinical trials in the same metric as for classical analysis. Psychological Methods, 14, 43-53. doi:10.1037/a0014699

16.

Flannery

D. J.

Montemayor

Eberly

Torquati

(1993). Unraveling the ties that bind: Affective expression and perceived conflict in parent adolescent interactions. Journal of Social and Personal Relationships, 10, 495-509. doi:10.1177/0265407593104002

17.

Garralda

M. E.

(2011). Unexplained physical complaints. Pediatric Clinics of North America, 58, 803-813. doi:10.1016/j.pcl.2011.06.002

18.

Garside

R. B.

Klimes-Dougan

(2002). Socialization of discrete negative emotions: Gender differences and links with psychological distress. Sex Roles, 47(3-4), 115-128. doi:10.1023/A%3A1021090904785%I

19.

Gilleland

Suveg

Jacob

M. L.

Thomassin

(2009). Understanding the medically unexplained: Emotional and familial influences on children’s somatic functioning. Child: Care, Health and Development, 35, 383-390. doi:10.1111/j.1365-2214.2009.00950.x

20.

Goldberg

D. P.

(1981). General Health Questionnaire. Winsor, UK: NFER Nelson.

21.

Gottman

J. M.

Katz

L. F.

Hooven

(1997). Meta-emotion: How families communicate emotionally. Mahwah, NJ: Lawrence Erlbaum.

22.

Gratz

K. L.

Roemer

(2004). Multidimensional assessment of emotion regulation and dysregulation: Development, factor structure, and initial validation of the Difficulties in Emotion Regulation Scale. Journal of Psychopathology and Behavioral Assessment, 26, 41-54. doi:10.1023/B:JOBA.0000007455.08539.94

23.

Havighurst

S. S.

Harley

A. E.

Kehoe

Pizarro

(2012). Tuning in to teens: Emotionally intelligent parenting (Program manual). Melbourne, Australia: The University of Melbourne.

24.

Havighurst

S. S.

Wilson

K. R.

Harley

A. E.

Prior

M. R.

(2009). Tuning in to kids: An emotion-focused parenting program—Initial findings from a community trial. Journal of Community Psychology, 37, 1008-1023. doi:10.1002/jcop.20345

25.

Havighurst

S. S.

Wilson

K. R.

Harley

A. E.

Prior

M. R.

Kehoe

(2010). Tuning in to kids: Improving emotion socialization practices in parents of preschool children—Findings from a community trial. Journal of Child Psychology and Psychiatry, 51, 1342-1350. doi:10.1111/j.1469-7610.2010.02303.x

26.

Heck

R. H.

Thomas

S. L.

Tabata

L. N.

(2010). Multilevel and longitudinal modeling with IBM SPSS: Quantitative methodology series. New York, NY: Routledge.

27.

Husain

Browne

Chalder

(2007). A review of psychological models and interventions for medically unexplained somatic symptoms in children. Child and Adolescent Mental Health, 12, 2-7. doi:10.1111/j.1475-3588.2006.00419.x

28.

Jellesma

F. C.

Rieffe

Terwogt

M. M.

Kneepkens

C. M. F.

(2006). Somatic complaints and health care use in children: Mood, emotion awareness and sense of coherence. Social Science & Medicine, 63, 2640-2648. doi:10.1016/j.socscimed.2006.07.004

29.

Jellesma

F. C.

Rieffe

Terwogt

M. M.

Westenberg

P. M.

(2008). Do parents reinforce somatic complaints in their children? Health Psychology, 27, 280-285. doi:10.1037/0278-6133.27.2.280

30.

Kehoe

C. E.

Havighurst

S. S.

Harley

A. E.

(2014). Tuning in to teens: Improving parent emotion socialization to reduce youth internalizing difficulties. Social Development, 23, 413-431. doi:10.1111/sode.12060

31.

Levy

R. L.

Whitehead

W. E.

Walker

L. S.

Von Korff

Feld

A. D.

Garner

Christie

(2004). Increased somatic complaints and health care utilization in children: Effects of parent IBS status and parent response to gastrointestinal symptoms. American Journal of Gastroenterology, 99, 1-10. doi:10.1111/j.1572-0241.2004.40478.x

32.

Magai

(1996). Emotions as a child. New York, NY: Long Island University.

33.

Magai

O’Neal

C. R.

(1997). Emotions as a child (Child version). Brooklyn, NY: Long Island University.

34.

Maliken

A. C.

Katz

L. F.

(2013). Exploring the impact of parental psychopathology and emotion regulation on evidence-based parenting interventions: A transdiagnostic approach to improving treatment effectiveness. Clinical Child and Family Psychology Review, 16, 173-186. doi:10.1007/s10567-013-0132-4

35.

Morris

A. S.

Silk

J. S.

Steinberg

Myers

S. S.

Robinson

L. R.

(2007). The role of the family context in the development of emotion regulation. Social Development, 16, 361-388. doi:10.1111/j.1467-9507.2007.00389.x

36.

Murray

D. M.

(1998). Design and analysis of group-randomized trials. New York, NY: Oxford University Press.

37.

Nauta

M. H.

Scholing

Rapee

R. M.

Abbott

Spence

S. H.

Waters

(2004). A parent report measure of children’s anxiety: Psychometric properties and comparison with child-report in a clinic and normal sample. Behaviour Research and Therapy, 42, 813-839. doi:10.1016/S0005-7967(03)00200-6

38.

O’Neal

C. R.

Magai

(2005). Do parents respond in different ways when children feel different emotions? The emotional context of parenting. Development and Psychopathology, 17, 467-487. doi:10.1017/S0954579405050224

39.

Pallant

(2007). SPSS survival manual: A step-by-step guide to data analysis using SPSS for Windows (3rd ed.). Crows Nest, Australia: Allen & Unwin.

40.

Pocock

S. J.

Assmann

S. E.

Enos

L. E.

Kasten

L. E.

(2002). Subgroup analysis, covariate adjustment and baseline comparisons in clinical trial reporting: Current practice and problems. Statistics in Medicine, 21, 2917-2930. doi:10.1002/sim.1296

41.

Preacher

K. J.

Hayes

A. F.

(2008). Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple mediator models. Behavior Research Methods, 40, 879-891. doi:10.3758/BRM.40.3.879

42.

Ralph

Sanders

M. R.

(2006). The “Teen Triple P” Positive Parenting Program: A preliminary evaluation. Youth Studies Australia, 25(2), 41-48. Available from http://www.acys.info/

43.

Raval

V. V.

Martini

T. S.

(2011). “Making the child understand”: Socialization of emotion in Urban India. Journal of Family Psychology, 25, 847-856. doi:10.1037/a0025240

44.

Raval

V. V.

Martini

T. S.

Raval

P. H.

(2010). Methods of, and reasons for, emotional expression and control in children with internalizing, externalizing, and somatic problems in urban India. Social Development, 19, 93-112. doi:10.1111/j.1467-9507.2008.00528.x

45.

Rieffe

Terwogt

M. M.

Bosch

J. D.

(2004). Emotion understanding in children with frequent somatic complaints. European Journal of Developmental Psychology, 1, 31-47. doi:10.1080/17405620344000013

46.

Rieffe

Terwogt

M. M.

Bosch

J. D.

Kneepkens

C. M. F.

Douwes

A. C.

Jellesma

F. C.

(2007). Interaction between emotions and somatic complaints in children who did or did not seek medical care. Cognition & Emotion, 21, 1630-1646. doi:10.1080/02699930701238495

47.

Saarni

(1999). The development of emotional competence. New York, NY: Guilford Press.

48.

Schulte

I. E.

Petermann

(2011). Familial risk factors for the development of somatoform symptoms and disorders in children and adolescents: A systematic review [Review]. Child Psychiatry & Human Development, 42, 569-583. doi:10.1007/s10578-011-0233-6

49.

Silk

J. S.

Shaw

D. S.

Prout

J. T.

O’Rourke

Lane

T. J.

Kovács

(2011). Socialization of emotion and offspring internalizing symptoms in mothers with childhood-onset depression. Journal of Applied Developmental Psychology, 32, 127-136. doi:10.1016/j.appdev.2011.02.001

50.

Spence

S. H.

(1998). A measure of anxiety symptoms among children. Behaviour Research and Therapy, 36, 545-566. doi:10.1016/S0005-7967(98)00034-5

51.

Spence

S. H.

Barrett

P. M.

Turner

C. M.

(2003). Psychometric properties of the Spence Children’s Anxiety Scale with young adolescents. Journal of Anxiety Disorders, 17, 605-625. doi:10.1016/S0887-6185(02)00236-0

52.

Steinhausen

H.-C.

Winkler Metzke

(2007). Continuity of functional-somatic symptoms from late childhood to young adulthood in a community sample. Journal of Child Psychology and Psychiatry, 48, 508-513. doi:10.1111/j.1469-7610.2006.01721.x

53.

Terwogt

M. M.

Rieffe

Miers

A. C.

Jellesma

F. C.

Tolland

(2006). Emotions and self-esteem as indicators of somatic complaints in children. Infant and Child Development, 15, 581-592. doi:10.1002/icd.479

54.

Walker

L. S.

Williams

S. E.

Smith

C. A.

(2006). Parent attention versus distraction: Impact on symptom complaints by children with and without chronic functional abdominal pain. Pain, 122, 43-52. doi:10.1016/j.pain.2005.12.020

55.

Warner

R. M.

(2012). Applied statistics: From bivariate through multivariate techniques. Thousand Oaks, CA: Sage.

56.

Weersing

V. R.

Rozenman

M. S.

Maher-Bridge

Campo

J. V.

(2012). Anxiety, depression, and somatic distress: Developing a transdiagnostic internalizing toolbox for pediatric practice. Cognitive and Behavioral Practice, 19, 68-82. doi:10.1016/j.cbpra.2011.06.002