Skip to main content
NCBI home page
Journal of Pediatric Psychology logoLink to Journal of Pediatric Psychology
. 2014 Sep 24;40(3):309–319. doi: 10.1093/jpepsy/jsu079

The Social Functioning of Siblings of Children With Cancer: A Multi-Informant Investigation

Melissa A Alderfer 1,✉, Caroline Stanley 2, Rowena Conroy 2, Kristin A Long 3, Diane L Fairclough 4, Anne E Kazak 5, Robert B Noll 3
PMCID: PMC4366447  PMID: 25256156

Abstract

Objective This study examined social functioning among siblings of children with cancer. Method A case–control design was applied to school- and home-based data from multiple informants (peers, teachers, mothers, and self). Social reputation and peer acceptance within the classroom was compared for 87 siblings (aged 8–16 years) and 256 demographically matched peers. Self-perceptions of peer relationships and parent-reported social competence were examined among 67 siblings and 67 matched comparisons.  Results Peer reports (N = 1,633) indicated no differences between siblings and comparisons for social reputation, number of friendships, reciprocated friendships, or peer acceptance. Self-reported prosocial behavior and teacher-reported likability were higher for siblings than comparisons. Self-reported loneliness, friendship quality, and perceived social support did not differ between groups. Mothers reported less involvement in activities and poorer school performance for siblings than comparisons. Conclusions Peer relationships of siblings of children with cancer are similar to classmates, though they experience small decrements in activity participation and school performance.

Keywords: childhood cancer, peer relationships, siblings, social functioning

In a pediatric health-care environment focused on family-centered care, it is important for pediatric psychologists to be aware of the psychological impact of childhood chronic illness on healthy siblings within the family. Meta-analyses indicate that, as a group, siblings of children with chronic illness experience modestly sized decrements in psychological adjustment when contrasted with norms and control groups (Sharpe & Rossiter, 2002; Vermaes, van Susante, & van Bakel, 2012). Comparisons across chronic conditions suggest that those at greatest risk for elevated distress are siblings of children with disorders that impact on daily functioning, require intrusive treatment regimens, and restrict the sibling in social realms (Sharpe & Rossiter, 2002). Siblings of children with cancer fall into this higher risk group.

Pediatric cancers are newly diagnosed in approximately 14,000 families in the United States each year (Howlader et al., 2013) and are the leading cause of death by disease for children (Heron et al., 2010). Siblings of children with cancer may witness their brother or sister in physical and emotional pain, experience sudden and extended separations from him or her, and worry that he or she will die (Alderfer, Labay, & Kazak, 2003). They also may be separated from parents for extended periods, have their daily routines disrupted, and need to negotiate changes in family roles and responsibilities due to cancer and its treatment (Alderfer & Kazak, 2006; Long & Marsland, 2011). Further, increased levels of parental distress (Vrijmoet-Wiersma et al., 2008), combined with the time demands of cancer treatment, may make it difficult for parents to attend fully to their healthy children (Enskar, Carlsson, Golsater, Hamrin, & Kreuger, 1997; Patterson, Holmes, & Gurney, 2004). A recent systematic review concluded that many siblings of children with cancer experience negative emotional reactions, poor quality of life, and academic difficulties in the months immediately following the cancer diagnosis; however, with time, most adjust well emotionally to the cancer experience (Alderfer et al., 2010).

Much less, however, is known about the social functioning and specifically the peer relationships of siblings of children with cancer. It is important to know whether cancer disrupts these aspects of the sibling’s life, because the peer social system contributes to child and adolescent development (Bronfenbrenner, 1979; Newcomb & Bagwell, 1995). Support from friends and classmates predicts emotional adjustment and resilience, with lower peer support associated with more anxiety, depression, and behavioral maladjustment among typically developing children (Demaray & Malecki, 2002; Demaray, Malecki, Davidson, Hodgson, & Rebus, 2005; Rubin, Bukowski, & Parker, 2006). Of particular relevance to families facing cancer, peer support has been found to compensate for lost parental support when families are disrupted (Stocker, 1994; Teja & Stolberg, 1993). Because of the significant role that peer relationships and the school context play in normative development and adjustment to challenges that disrupt the family, problems with social functioning for siblings of children with cancer may be particularly detrimental. Alternatively, intact social functioning may be an important resource for siblings and a sign of resilience.

A complete evaluation of the social functioning of children is best accomplished across school and home contexts using data from multiple sources. Peer nomination procedures in the classroom are considered a gold standard for assessing social functioning, as they are reliable and valid measures that predict future functioning (Bukowski, Cillessen, & Velasquez, 2012; Parker & Asher, 1987). While reports of social functioning of children based on self-report, teacher report, or parent report tend to show low correspondence with peer nominations (Noll & Bukowski, 2012), it is nonetheless important to consider these non-peer perspectives as part of a comprehensive multi-informant assessment of child social functioning. Teachers and parents may observe different behaviors of the same child in different contexts (Kraemer et al., 2003), and some important aspects of social functioning (e.g., social support) are best assessed from the perspective of the target child.

To date, with few exceptions (Labay & Walco, 2004), studies reporting on the social functioning of siblings of children on cancer treatment have been qualitative (Freeman, O’Dell, & Meola, 2000; Murray, 2002; Prchal & Landolt, 2012; Sidhu, Passmore, & Baker, 2005; Sloper, 2000). These studies suggest that siblings of children with cancer have decreased opportunities for social encounters and while friends are an important source of support and distraction for siblings, relationships can be disrupted by attention toward the child with cancer. The purpose of the current study was to provide a more comprehensive quantitative examination of the social functioning of siblings of children with cancer by contrasting them with matched comparison peers across a range of measures completed by multiple informants (peers, teachers, parents, and self) in the school and home settings. We tested the hypothesis that siblings of children with cancer would have poorer social functioning than comparison peers across all measures.

Methods

General Overview

This study included two phases. Phase 1 data were collected in the classrooms of siblings and included the peer nomination measures and teacher ratings. Phase 2 data, including parent report and self-report measures, were collected in the homes of siblings (i.e., “target families”) and, for each sibling, one matched comparison. Institutional review board approval was granted for all procedures. School-based approval procedures (e.g., superintendent or school board approval) were also completed as needed before Phase 1 data collection.

Participants and Recruitment Procedures

Eligible target families had: (a) a living child either on treatment for cancer or off-treatment but within 2 years of diagnosis; (b) a child without a chronic illness aged 8–16 years who attended regular classes (no full-time special education or homeschooling); and (c) English fluency. In families with multiple eligible siblings, the sibling closest in age to the child with cancer was chosen for participation. Detailed recruitment and enrollment numbers are provided in Figure 1.

Figure 1.

Figure 1.

Open in a new tab

Recruitment and enrollment flow chart.

Potentially eligible target families were identified from the tumor registry of a large northeastern children’s hospital. They were sent a letter of invitation and then received a follow-up phone call to ascertain eligibility and interest in participating in a study of “family and peer influences on sibling adjustment to childhood cancer.” The enrollment rate of eligible target families was 80% for Phase 1. Target siblings’ schools were then contacted; 77% agreed to distribute recruitment letters and consent forms in the siblings’ classrooms. The project was described to students as “a study of friendships during times of stress.” No mention was made of the target child, the hospital, or our interest in siblings of children with cancer. This was done to avoid stigmatizing the sibling. Data were collected in 87 classrooms; 88% of students provided consent/assent. On average, Phase 1 data collection occurred 15.4 months (SD = 7.3, range: 3–34) after the cancer diagnosis. The distribution of cancer diagnoses was 40% leukemia/lymphoma, 46% solid, and 14% brain tumors. Target siblings’ classmates of the same gender and closest race/ethnicity were identified. From these, the three closest in age to the sibling were selected as classroom comparisons. In five classes, only two comparison children wereavailable, resulting in 256 comparison classmates (Table I).

Table I.

Demographics

 Siblings  Comparisons
Phase 1: School sample N = 87 N = 256
    Age: M (SD) in years 11.7 (2.3) 11.6 (2.3)
    Female gender 57% (n = 50) 57% (n = 157)
    Race/Ethnicity
        White/non-Hispanic 76% (n = 66) 78% (n = 201)
        Black/non-Hispanic 15% (n = 13) 13% (n = 33)
        White/Hispanic 5% (n = 4) 6% (n = 15)
        Other/Unknown 5% (n = 4) 3% (n = 7)
Phase 2: Home sample N = 67 N = 67
    Age: M (SD) in years 12.2 (2.4) 12.3 (2.4)
    Female gender 55% (n = 37)a 55% (n = 37)
    Race/ethnicity
        White/non-Hispanic 81% (n = 54)b 85% (n = 57)c
        Black/non-Hispanic 13% (n = 9) 13% (n = 9)
        White/Hispanic 3% (n = 2) 2% (n = 1)
        Other/Unknown 2% (n = 1)
    Mother’s education (median category) University graduate University graduate
    Household income (median category) $100,000–$124,999 $75,000–$99,999d
    Married/partnered parents 85% (n = 57) 82% (n = 54)e
Open in a new tab

Note. aNot different from Phase 1, χ2 = 0.01, p = .91.

bNot different from Phase 1, χ2 = 0.47, p = .49.

cNot different from sibling sample, McNemar, p = .56.

dNot different from sibling sample, Wilcoxon signed-rank test, p = .95.

eNot different from sibling sample, McNemar, p = .75.

After Phase 1 school data collection was complete, each target family was contacted for Phase 2 participation. Of those eligible families who were able to be contacted, 93% completed a second informed consent process and enrolled (Figure 1). For each sibling enrolled in Phase 2, a single comparison classmate was sought by contacting the matched peers identified during Phase 1, beginning with the classmate closest in age to the sibling, and continuing until an eligible family agreed to participate. Comparison families were told that they were selected for participation because their child had a specific set of demographic characteristics and experiences (i.e., “similar in some ways, but different in others”) that, when compared with another family’s responses, would help in understanding the role of family and friendships during times of stress. Comparison families were ineligible if they included a child with a life-threatening illness. Of the contacted eligible comparison families, 89% enrolled, providing 67 matched pairs for analysis. All but three comparison classmates were within the three closest age matches identified and used in Phase 1. As in past research (Noll et al., 2007), this method resulted in pairs that were well-matched on demographic characteristics (Table I). The 20 siblings for whom home visit data were not available, or not analyzed owing to missing comparisons, were no different from the 67 for whom data were available and analyzed in terms of ethnicity/race (White/non-Hispanic vs. other: χ2 [1, N = 85] = 1.3, p = .26), gender (χ2 [1, N = 85] = 0.6, p = .44), age (t = 0.31, p = .75) or any of the classroom-based measures (t values < 1.65, p values > .10).

Classroom Data Collection Procedures and Measures

Once approximately 80% of consent forms had been returned for a classroom, the school-based data collection was scheduled. A team of research assistants visited the classroom, explained the project to the students and collected data using procedures developed, tested, and described in prior work (Noll et al., 1999). Students and teachers completed the following peer relationship measures.

Revised Class Play

The Revised Class Play (RCP; Zeller, Vannatta, Schafer & Noll, 2003) asks students to imagine that they are directing a play and must ‘‘cast’’ members of their class into 30 hypothetical ‘‘roles’’ presented as individual items. The roles capture different social styles (e.g., sensitive-isolated: “someone who would rather play alone than with others”; aggressive-disruptive: “someone who gets into a lot of fights”). Respondents may nominate only one student per role/item, but can nominate the same person for multiple roles; they cannot cast themselves. To avoid gender role stereotyping, only classmates of the same gender as the target child can be nominated. The number of peer nominations that each child receives for each role/item is tallied and converted to a z-score (M = 0, SD = 1) to adjust for differences in the possible number of nominations (i.e., unequal class sizes). The resulting item scores are summed to form four subscales: Leadership-Popularity, Prosocial, Aggressive-Disruptive, and Sensitive-Isolated. In our sample, internal consistency across subscales ranged from .85 to .90.

After completing the peer nomination version of the RCP, each student rated how well they could perform each role on a 4-point scale (1 = I could never play this part to 4 = I would be great at this part). Means were calculated across items contributing to each of the four dimensions listed above and were standardized within gender for each class. Internal consistency across subscales ranged from .72 to .80 in our sample.

Best Friend Nominations

This measure asks each participant to nominate their three best friends from among their classmates. Two scores are derived for each student: (a) the total number of friendship nominations received and (b) the number of reciprocated friendships. Both scores are stable and valid indicators of peer relations (Bukowski & Hoza, 1989; Gottman, Gonso, & Rasmussen, 1975) and are standardized to z-scores within gender for each class to allow for analysis.

Peer Acceptance Ratings

This measure asks students to rate how much they like each of their classmates on a 5-point scale (1 = someone you do not like to 5 = someone you like a lot). Mean acceptance ratings are calculated across respondents for each student to provide an assessment of their social acceptance within the classroom. Scores are standardized within gender for each class. This measure has shown stability across a 4-week interval (Asher, Singleton, Tinsley, & Hymel, 1979; Ladd, 1981).

Pupil Evaluation Inventory – Teacher Version

This measure asks teachers to indicate which of their students are characterized by each of 34 descriptors (e.g., “helps others”). The teacher can choose multiple students, or none, for each item. Items load onto three subscales validated by factor analysis: Aggression, Withdrawal, and Likeability. The number of endorsements that each child receives across the items loading onto each subscale is calculated. In past research, teacher and peer ratings on the Pupil Evaluation Inventory – Teacher Version (PEI) have been significantly correlated, demonstrating validity (Pekarik, Prinz, Liebert, Weintraub, & Neale, 1976). For our sample, internal consistencies were: Aggression, α = .87; Withdrawal, α = .73; and Likeability, α = .68.

Home Visit Data Collection Procedures and Measures

Two research assistants traveled to the home of each Phase 2 target and comparison family to gather consent/assent and collect data. Families received $50 for the home visit. Parents and children completed the following measures of child social functioning.

Child Behavior Checklist

This checklist gathers parents’ ratings of a child’s competencies and behavioral/emotional problems. The social competence section assesses the child’s involvement with friends, participation in activities, and performance in school, and results in a total social competence score. Mothers’ responses to this portion of the Child Behavior Checklist (CBCL) were analyzed for this article. The three subscale scores (Social, Activities, School) and the summary score are converted to T-scores (M = 50, SD = 10) based on child age and gender. Lower scores indicate poorer functioning and are classified as follows: Clinical (T-score ≤ 30), Borderline Clinical (T-score = 31–35), and Normal (T-score > 35). Test–retest reliability of these scales is high over a period averaging 8 days (r values = .82–.93) and internal consistencies are adequate (α values = .63–.79; Achenbach & Rescorla, 2001). In our sample, subscale alphas were as follows: Activities, α = .61; Social, α = .60; and School, α = .56.

Loneliness Questionnaire

The Loneliness Questionnaire (LQ) is a 24-item measure with 16 items assessing loneliness, social inadequacy, and perceived social status at school (e.g., “I’m lonely at school”), plus eight filler items (e.g., “I like to read”). Respondents indicate how true each item is for them on a 5-point scale (1 = always true about me to 5 = not true at all about me). After recoding of some items, higher summed scores across the 16 content items reflect greater loneliness. Loneliness scores are significantly inversely correlated with peer acceptance ratings and number of friendship nominations (Asher & Wheeler, 1985). In our sample, internal consistency was .89.

Child and Adolescent Social Support Scale

This 60-item measure assesses child and adolescent perceptions of social support from parents, teachers, classmates, close friends, and people at school. Children indicate on a 6-point scale (1 = never to 6 = always) how often each specified support action occurs from each of these general groups of people, then how important that action is to them on a 3-point scale (1 = not very important to 3 = very important). Frequency and importance ratings are summed separately. In past research, test–retest reliability over 8–10 weeks indicated moderate to high consistency (r values = .45–.78; Malecki, Demaray, & Elliott, 2004). Convergent validity has been established with other social support measures (Malecki & Demaray, 2002). Current analyses used the classmates, close friend, and people at school social support (frequency) subscales. Internal consistency in our sample across these subscales ranged from .93 to .95.

McGill Friendship Questionnaire – Child Version

The McGill Friendship Questionnaire – Child Version (MFQ; Aboud, Mendelson, & Purdy, 2003) measures specific qualities of friendships. The respondent names his/her three best friends and then rates those friendships across 30 items contributing to six subscales: Reliable Alliance (e.g., “Would still want to be my friend even if we had a fight.”), Emotional Security (e.g., “Makes me feel better when I am upset.”), Companionship (e.g., “Is enjoyable to be with.”), Assistance (e.g., “Helps me when I need it.”), Intimacy (e.g., “Is someone I can tell secrets to.”), and Self-Validation (e.g., “Makes me feel sure of myself.”). For this study, respondents reported how often each item occurred within each of their friend relationships using a 5-point scale (0 = never to 4 = always). Internal consistency in our sample across subscales ranged from .79 to .86.

Analytical Approach

Classroom Data

Hierarchical linear mixed models (Jennrich & Schluchter, 1986) were used to examine differences between target siblings and comparison classmates, allowing for the nesting of children within classroom. With 87 target siblings and 256 comparison classmates, statistical power was > 90% across all classroom analyses to detect moderately sized effects (d = .5). To detect small effects (d = .3), power was 52–69% across the peer-report scales, 54–65% across the self-report scales, and 71–78% across the teacher-report scales.

Home Visit Data

Multivariate analyses of variance (MANOVAs) with child (siblings, comparison) as a within-subjects variable and each measures’ (CBCL, Child and Adolescent Social Support Scale [CASSS], MFQ) subscales as the dependent variables or paired t tests (LQ) were used to contrast siblings and comparison peers. For the MFQ, friend (best friend, friend 2, friend 3) was also a within-subjects variable. To fully explore the data, MANOVAs were followed up with univariate analyses with Sidak adjustment to conserve alpha. The least powerful of these analyses (paired t test) had 94% power to detect small effects (d = .3). The CASSS required a reflection and square root procedure to correct skew. The MFQ required computation of the inverse of the data after reflection (Tabachnick & Fidell, 1996); one subscale (Reliable Alliance of Best Friend) remained significantly skewed, but the pattern was consistent across the two groups. To characterize the sample, descriptive statistics (Mean [M], standard deviation [SD]) were calculated for the T-scores on each CBCL scale, along with the percentage of siblings and comparison peers falling outside the normal range.

Results

Social Reputation, Best Friend Nominations, and Acceptance: Peer Report, Self-Report and Teacher Report

Classroom-based peer-reported data indicated no differences in social functioning between siblings and comparison classmates (Table II). Peer reports of social behavior/reputation on the RCP demonstrated no differences in rate of nominations for siblings and comparison children on any of the four dimensions. There were no significant group differences in total best friend nominations or number of reciprocated friendships. Finally, peer acceptance ratings were not significantly different for siblings versus comparison classmates.

Table II.

Social Reputation, Friend Nominations, and Acceptance: Siblings Versus Comparison Peers

Siblings M (SD) Comparisons M (SD) F p d
Peer ratings
    Leadership-popularity 0.08 (1.0) 0.07 (0.92) 0 .95 .01
    Prosocial 0.04 (0.93) 0.01 (1.0) 0.08 .78 .03
    Aggressive-disruptive 0.02 (1.06) −0.01 (0.87) 0.09 .76 .04
    Sensitive-isolated −0.04 (0.91) −0.07 (0.86) 0.04 .84 .03
    Total best friend nominations 0.07 (0.96) 0.03 (0.98) 0.08 .77 .04
    Reciprocated best friends −0.02 (0.91) 0.02 (0.96) 0.12 .73 −.05
    Peer acceptance 0.02 (0.93) 0.02 (0.93) 0 1.0 0
Self-ratings
    Leadership-popularity 0.18 (0.89) 0.01 (0.91) 1.69 .20 .19
    Prosocial 0.16 (0.80) −0.11 (0.90) 6.70 .01 .32
    Aggressive-disruptive −0.01 (0.99) 0.02 (0.95) 0.06 .81 −.04
    Sensitive-isolated −0.02 (0.94) −0.11 (0.91) 0.62 .43 .10
Teacher ratings
    Aggression 1.26 (2.19) 1.31 (2.40) 0.04 .85 −.02
    Withdrawal 0.78 (1.36) 0.61 (1.17) 1.37 .24 .14
    Likeability 2.18 (1.66) 1.74 (1.61) 6.11 .02 .27
Open in a new tab

Note. Bolded values are statistically significant.

Self-ratings by siblings and comparison classmates on the RCP did not differ in regard to the Leadership-Popularity, Aggressive-Disruptive, or Sensitive-Isolated dimensions. However, siblings rated themselves higher on the Prosocial scale. Similarly, teachers rated siblings and comparison classmates equally on the Aggressive and Withdrawal subscales of the PEI, but rated siblings as more Likeable than comparison classmates.

School Performance, Engagement in Activities, and Social Functioning: Mothers’ Reports

There were significant, but small, differences between siblings and comparison children on the parent-report CBCL. MANOVA, using raw scores, revealed a statistically significant effect across the three social subscales (Wilks λ = .86; F(3,57) = 3.14, p = .03, partial η2 = .14). Follow-up univariate analyses (Table III) revealed no differences between siblings and comparison children on the social subscale. However, parents rated target siblings as less involved in activities and performing poorer at school than comparison peers. The size of these significant effects was small. T-scores and the percentage of participants falling outside the normal range on each scale are summarized in Table III.

Table III.

Social Competence Scores on the Child Behavior Checklist: Siblings (n = 60)a Versus Comparison Peers (n = 60)a

T-scoresb
 Classification outside normal rangec,d
 Statistical comparison of raw scores
Siblings M (SD) Peers M (SD) Siblings % (n) Peers % (n) Siblings M (SE) Peers M (SE) F (1, 59) p d
Total Social Competence 47.7 (10.8) 52.1 (11.2) 15 (9) 8 (5) 23.76 (0.61) 25.60 (0.57) 6.55 .01 .100
    Activities 44.7 (9.2) 48.7 (9.3) 13 (8) 7 (4) 9.83 (0.34) 10.78 (0.30) 1.61 .03 .078
    Social 50.1 (9.4) 52.0 (9.6) 8 (5) 10 (6) 8.84 (0.34) 9.39 (0.34) 5.0 .21 .027
    School 48.9 (7.2) 51.2 (6.3) 3 (2) 3 (2) 5.09 (0.14) 5.43 (0.12) 6.11 .02 .094
Open in a new tab

Note. aFamilies in which only fathers completed the CBCL and incomplete sibling-comparison pairs were dropped from CBCL analyses.

bThe mean in the normative sample is 50 with a standard deviation of 10. Lower scores indicate poorer functioning.

cThe percentage of children expected outside the normal range for the Total Social Competence Scale is 16% and for the subscales is 8%.

dTwenty-seven of the siblings reported on here were included in the sample analyzed in Alderfer and Hodges, 2010.

Loneliness, Social Support, and Relationship Quality: Sibling and Comparison Self-Reports

Siblings (M = 35.78; SD = 12.13) and comparison peers (M = 32.85; SD = 13.43) did not differ significantly on reports of loneliness within the classroom (t(66) = 1.28, p = .21) or social support across sources (Wilks λ = .94; F(3, 60) = 1.21, p = .31, partial η2 = .06). Follow-up univariate comparisons (Table IV) indicated no significant differences between siblings and comparisons in amount of support from friends, classmates, or others in school. Siblings’ ratings of friendship quality were no different from comparison peers’ ratings (Wilks λ = .91; F(6, 59) = .98, p = .45, partial η2 = .09). Follow-up univariate analyses (Table IV) revealed no differences between siblings and comparisons on any friendship quality subscale.

Table IV.

Perceived Social Support and Relationship Quality, Siblings Versus Comparison Peers

Sibling M (SE) Comparison M (SE) F p Partial η2
Perceived social support (CASSS)a
    Close friend 2.90 (0.20) 2.48 (0.17) 2.62 .11 .041
    Classmates 3.92 (0.22) 3.60 (0.20) 1.48 .23 .023
    Others at school 3.92 (0.23) 3.87 (0.22) 0.04 .85 .001
Relationship Quality (MFQ)b
    Companionship 0.70 (0.04) 0.68 (0.03) 0.22 .65 .003
    Validation 0.60 (0.04) 0.64 (0.04) 0.53 .47 .008
    Assistance 0.58 (0.04) 0.58 (0.04) 0.00 .97 .000
    Intimacy 0.57 (0.04) 0.55 (0.04) 0.11 .74 .002
    Emotional security 0.67 (0.04) 0.64 (0.04) 0.24 .63 .004
    Reliability of alliance 0.74 (0.04) 0.72 (0.04) 0.14 .71 .002
Open in a new tab

Note. aChild and Adolescent Social Support Scale; transformed scores have a possible range of 1–7.8, with lower scores indicating more support.

bMcGill Friendship Questionnaire, transformed scores have a possible range of .05–1.0, with higher scores indicating better relationship quality.

Discussion

This study contributes significantly to our knowledge of the social functioning of siblings of children with cancer and indicates that overall, siblings’ peer relationships are strikingly similar to matched classroom peers. Through a methodologically rigorous case–control design with adequate power, social functioning was assessed in this study through the perspectives of multiple informants. Findings indicate that siblings are similar to comparison peers in regard to social behavior, number of friends, number of reciprocated friendships, and level of peer acceptance. Siblings’ self-perceptions of their social adequacy at school; social support from friends, classmates, and others at school; and relationship quality with friends are also similar to comparison peers. Siblings rated themselves higher in prosocial behavior than did matched-comparisons, and teachers rated siblings as more likeable than classroom controls. These findings do not support the hypothesis that siblings’ social functioning is poorer than peers.

These findings do suggest that despite the challenges of childhood cancer (e.g., disruptions to activities and school performance), siblings of children with cancer are socially resilient. These findings are encouraging; these social resources may help siblings overcome cancer-related emotional and practical challenges. In previous work (Alderfer & Hodges, 2010), support from friends, classmates, and others at school was negatively associated with symptoms of depression among siblings of children with cancer.

Siblings in the current study rated themselves as more prosocial, and teachers rated siblings as more likeable, than comparison peers. The sizes of these effects were small, but consistent, across informants and similar to teachers’ ratings of bereaved siblings (Gerhardt et al., 2012). While the teacher ratings may be biased, influenced by knowledge of which student was the target of our study and why, it is also possible that they and the siblings are reporting on subtle patterns in the siblings’ behavior that peers do not detect. The items on these subscales capture a constellation of characteristics (e.g., polite, helpful, fair, nice) that may reflect the increased maturity, empathy, and compassion that siblings show in response to the cancer experience (Freeman et al., 2000; Heffernan & Zanelli, 1997; Sloper, 2000).

Less involvement in activities and poorer school performance have been noted in previous qualitative and quantitative work regarding siblings of children with cancer (Freeman et al., 2000; Houtzager, Grootenhuis, Hoekstra-Webers, & Last, 2005; Labay & Walco, 2004; Lahteenmaki, Sjoblom, Korhonen, & Salmi, 2004; Murray, 2002; Prchal & Landolt, 2012; Sidhu et al., 2005; Sloper, 2000). What is noteworthy in our current findings is that these changes were of small magnitude and did not have a significant adverse impact on siblings’ peer relationships. These mother-reported difficulties may seem to contradict the positive report by the siblings and teachers regarding siblings’ prosocial behavior; however, these constructs represent different aspects of social functioning and could easily coexist. The more negative report of the mothers of siblings may also reflect, in part, their own level of cancer-related distress (De Los Reyes & Kazdin, 2005). Overall, on average, siblings’ peer relationships are similar to those of matched comparison peers in terms of number of friends, number of reciprocated friendships, level of peer acceptance, and self-perceived social support and friendship quality. Similar findings have been reported for siblings of children with sickle cell disease (Noll et al., 1995).

The findings of this study need to be considered within the context of its limitations. The sample was primarily White/non-Hispanic, well-educated, and high in socioeconomic status (SES). The sample reflects the demographic characteristics of the Oncology service from which families were drawn, and carefully matched comparisons were used for analysis. However, the sample may not represent or generalize to the broader population of families of children with cancer or specifically those of lower SES. Despite restricting the sample in regard to age and time since diagnosis, there was still heterogeneity in these variables and other treatment-related variables (e.g., intensity of treatment), and sibling social functioning may vary as a function of these variables as well as gender (Gerhardt et al., 2012). The study was not designed to examine differences among these various subgroups of siblings. The internal consistency of the CBCL in our sample, specifically the School subscale, was somewhat low, reducing statistical power (Bacon, 2004); however, our sample size and matched design provided enough power to uncover as statistically significant even small differences between the groups on this measure. Finally, the measures used did not specifically assess negative relationship qualities (i.e., conflict) or changes in friendships across time. Siblings may lose or “weed out” friendships that are weaker or less supportive while strengthening or establishing new friendships that are more satisfying and helpful in the face of dealing with cancer. More in-depth qualitative work may be useful to understand possible nuances in siblings’ friendships after cancer diagnosis. Similarly, longitudinal and mixed-methods designs could examine whether such relationship characteristics or changes are detrimental to or foster long-term sibling adjustment.

The results of this study are particularly important to pediatric psychologists working within family-centered care contexts or taking a family systems approach where the bidirectional relationships between family and illness are recognized. As part of the family, siblings are affected by the changes and stressors that result from childhood illnesses and often contribute to illness management activities within the family. The current study of sibling social functioning addresses important, but understudied, aspects of sibling adjustment and development: The peer and school contexts. By addressing this gap (e.g., it is the first to provide data from peers) and using a methodologically rigorous design, it contributes substantially to the existing literature regarding the social functioning of siblings of children with cancer, and the findings are encouraging. The peer relationships of siblings may serve as an important resource to them as they adapt to the changes that childhood illness brings to the family. Future work should investigate whether peer acceptance relates to sibling emotional adjustment and whether positive peer relationships moderate the impact of potentially disrupted family functioning on sibling adjustment to cancer. Larger samples are needed in future work to enable examination of the factors that contribute to social resilience and to identify variables that predict which siblings may be at risk for social functioning difficulties.

Acknowledgments

The authors thank the schools and families who participated and the research team including Rebecca Shorter, Shelby Rosario, Willy Tang, Lindsey Teitinen, Leela Jackson, Joanna Cohen, Jilda Hodges, David Moore, Kim Wesley, Stephanie Pelligra, Jessica Rosskam, Elizabeth Weyman, Chelsea Pasmore, Margaret Cheng, Rachel Bobrick, Ivy Pete, Kristen Craig, and Tracy Hills.

Funding

This work was supported by the American Cancer Society [MRSG05213 to M.A.A.] and the National Cancer Institute [K05CA128805 to A.E.K.].

Conflicts of interest: Dr. Fairclough currently receives grant funding from Biogen and Novartis.

References

  1. Aboud F E, Mendelson M J, Purdy K T. Cross-race peer relations and friendship quality. International Journal of Behavioral Development. 2003;27:165–173. [Google Scholar]
  2. Achenbach T M, Rescorla L A. Manual for the ASEBA school-age forms and profiles. Burlington, VT: University of Vermont; 2001. [Google Scholar]
  3. Alderfer M A, Hodges J. Supporting siblings of children with cancer: A need for family-school partnerships. School Mental Health. 2010;2:72–81. doi: 10.1007/s12310-010-9027-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Alderfer M A, Kazak A E. Family issues when a child is on treatment for cancer. In: Brown R T, editor. Comprehensive handbook of childhood cancer and sickle cell disease: A biopsychosocial approach. New York: Oxford University Press; 2006. pp. 53–74. [Google Scholar]
  5. Alderfer M A, Labay L, Kazak A E. Does posttraumatic stress apply to siblings of childhood cancer survivors? Journal of Pediatric Psychology. 2003;21:281–286. doi: 10.1093/jpepsy/jsg016. [DOI] [PubMed] [Google Scholar]
  6. Alderfer M A, Long K A, Lown E A, Marsland A L, Ostrowski N L, Hock J M, Ewing L J. Psychosocial adjustment of siblings of children with cancer: A systematic review. Psycho-oncology. 2010;19:789–805. doi: 10.1002/pon.1638. [DOI] [PubMed] [Google Scholar]
  7. Asher S R, Singleton L C, Tinsley B R, Hymel S. A reliable sociometric measure for preschool children. Developmental Psychology. 1979;15:443–444. [Google Scholar]
  8. Asher S, Wheeler V. Children's loneliness: A comparison of rejected and neglected peer status. Journal of Consulting and Clinical Psychology. 1985;53:500–505. doi: 10.1037//0022-006x.53.4.500. [DOI] [PubMed] [Google Scholar]
  9. Bacon D R. The contributions of reliability and pretests to effective assessment. Practical Assessment, Research and Evaluation. 2004;9(3) Retrieved from http://PAREonline.net/getvn.asp?v=9&n=3 (accessed 12 August 2014) [Google Scholar]
  10. Bronfenbrenner U. The ecology of human development. Cambridge: Harvard University Press; 1979. [Google Scholar]
  11. Bukowski W M, Cillessen A H N, Velasquez A M. The use of peer ratings in developmental research. In: Laursen B, Little T, Card N, editors. Handbook of developmental research methods. New York: Guilford; 2012. pp. 211–228. [Google Scholar]
  12. Bukowski W M, Hoza B. Popularity and friendship: Issues in theory, measurement, and outcomes. In: Berndt T, Ladd G, editors. Contributions of peer relationships to children’s development. New York: Wiley; 1989. pp. 15–45. [Google Scholar]
  13. De Los Reyes A, Kazdin A E. Informant discrepancies in the assessment of childhood psychopathology: A critical review, theoretical framework, and recommendations for further study. Psychological Bulletin. 2005;131:483–509. doi: 10.1037/0033-2909.131.4.483. [DOI] [PubMed] [Google Scholar]
  14. Demaray M K, Malecki C K. Critical levels of perceived social support associated with student adjustment. School Psychology Quarterly. 2002;17:213–241. [Google Scholar]
  15. Demaray M K, Malecki C K, Davidson L M, Hodgson K K, Rebus P J. The relationship between social support and student adjustment: A longitudinal analysis. Psychology in Schools. 2005;42:691–706. [Google Scholar]
  16. Enskar K, Carlsson M, Golsater M, Hamrin E, Kreuger A. Parental reports of changes and challenges that result from parenting a child with cancer. Journal of Pediatric Oncology Nursing. 1997;14:156–163. doi: 10.1016/s1043-4542(97)90051-5. [DOI] [PubMed] [Google Scholar]
  17. Freeman K, O’Dell C, Meola C. Issues in families of children with brain tumors. Oncology Nursing Forum. 2000;27:843–848. [PubMed] [Google Scholar]
  18. Gerhardt C A, Fairclough D L, Grossenbacher J C, Barrera M, Gilmer M J, Foster T L, Compas B E, Davies B, Hogan N S, Vannatta K. Peer relationships of bereaved siblings and comparison classmates after a child's death from cancer. Journal of Pediatric Psychology. 2012;37:209–219. doi: 10.1093/jpepsy/jsr082. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Gottman J, Gonso J, Rasmussen B. Social interaction, social competence, and friendship in children. Child Development. 1975;46:709–718. [PubMed] [Google Scholar]
  20. Heffernan S M, Zanelli A S. Behavior changes exhibited by siblings of pediatric oncology patients: A comparison between maternal and sibling description. Journal of Pediatric Oncology Nursing. 1997;14:3–14. doi: 10.1177/104345429701400102. [DOI] [PubMed] [Google Scholar]
  21. Heron M, Sutton P D, Xu J, Venture S J, Strobino D M, Guyer B. Annual summary of vital statistics: 2007. Pediatrics. 2010;125:4–15. doi: 10.1542/peds.2009-2416. [DOI] [PubMed] [Google Scholar]
  22. Houtzager B, Grootenhuis M, Hoekstra-Weebers J, Last B. One month after diagnosis: Quality of life, coping and previous functioning in siblings of children with cancer. Child: Care, Health and Development. 2005;31:75–87. doi: 10.1111/j.1365-2214.2005.00459.x. [DOI] [PubMed] [Google Scholar]
  23. Howlader N, Noone A M, Krapcho M, Garshell J, Neyman N, Altekruse S F, Kosary C L, Yu M, Ruhl J, Tatalovich Z, Cho H, Mariotto A, Lewis D R, Chen H S, Feuer E J, Cronin K A, editors. SEER Cancer Statistics Review. National Cancer Institute, Bethesda, MD; 1975-2010. Retrieved from http://seer.cancer.gov/csr/1975_2010/, based on November 2012 SEER data submission, posted to the SEER web site, April 2013. [Google Scholar]
  24. Jennrich R I, Schluchter M D. Unbalanced repeated-measures models with structured covariance matrices. Biometrics. 1986;42:805–820. [PubMed] [Google Scholar]
  25. Kraemer H C, Measelle J R, Ablow J C, Essex M J, Boyce W T, Kupfer D J. A new approach to integrating data from multiple informants in psychiatric assessment and research: Mixing and matching contexts and perspectives. American Journal of Psychiatry. 2003;160:1566–1577. doi: 10.1176/appi.ajp.160.9.1566. [DOI] [PubMed] [Google Scholar]
  26. Labay L, Walco G. Brief report: Empathy and psychological adjustment in siblings of children with cancer. Journal of Pediatric Psychology. 2004;29:309–314. doi: 10.1093/jpepsy/jsh032. [DOI] [PubMed] [Google Scholar]
  27. Ladd G. Effectiveness of a social learning method for enhancing children’s social interaction and peer acceptance. Child Development. 1981;52:171–178. [PubMed] [Google Scholar]
  28. Lahteenmaki P, Sjoblom J, Korhonen T, Salmi T. The siblings of childhood cancer patients need early support: A follow up study over the first year. Archives of Disease in Childhood. 2004;89:1008–1013. doi: 10.1136/adc.2002.012088. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Long K A, Marsland A L. Family adjustment to childhood cancer: A systematic review. Clinical Child and Family Psychology Review. 2011;14:57–88. doi: 10.1007/s10567-010-0082-z. [DOI] [PubMed] [Google Scholar]
  30. Malecki C K, Demaray M K. Measuring perceived social support: Development of the child and adolescent social support scale (CASSS) Psychology in Schools. 2002;39:1–18. [Google Scholar]
  31. Malecki C K, Demaray M K, Elliott S N. A working manual of the development of the child and adolescent social support scale. 2004. Unpublished manuscript, Northern Illinois University, Illinois. [Google Scholar]
  32. Murray J. A qualitative exploration of psychosocial support for siblings of children with cancer. Journal of Pediatric Nursing. 2002;17:327–337. doi: 10.1053/jpdn.2002.127177. [DOI] [PubMed] [Google Scholar]
  33. Newcomb A F, Bagwell C L. Children’s friendship relations: A meta-analytic review. Psychological Bulletin. 1995;117:306–347. [Google Scholar]
  34. Noll R B, Bukowski W M. Commentary: Social competence in children with chronic illness: The devil is in the details. Journal of Pediatric Psychology. 2012;37:959–966. doi: 10.1093/jpepsy/jss058. [DOI] [PubMed] [Google Scholar]
  35. Noll R B, Gartstein M A, Vannatta K, Correll J, Bukowski W M, Davies W H. Social, emotional, and behavioral functioning of children with cancer. Pediatrics. 1999;103:71–78. doi: 10.1542/peds.103.1.71. [DOI] [PubMed] [Google Scholar]
  36. Noll R B, Reiter-Purtill J, Vannatta K, Gerhardt C, Short A. Peer relationships and emotional well-being of children with sickle cell disease: A controlled replication. Child Neuropsychology. 2007;13:173–187. doi: 10.1080/09297040500473706. [DOI] [PubMed] [Google Scholar]
  37. Noll R B, Yosua L A, Vannatta K K, Kalinyak K, Bukowski W M, Davies W H. Social competence of siblings of children with Sickle Cell Anemia. Journal of Pediatric Psychology. 1995;20:165–172. doi: 10.1093/jpepsy/20.2.165. [DOI] [PubMed] [Google Scholar]
  38. Parker J G, Asher S R. Peer relations and later personal adjustment: Are low-accepted children at risk? Psychological Bulletin. 1987;102:357–389. doi: 10.1037//0033-2909.102.3.357. [DOI] [PubMed] [Google Scholar]
  39. Patterson J M, Holm K E, Gurney J G. The impact of childhood cancer on the family: A qualitative analysis of strains, resources, and coping behaviors. Psycho-oncology. 2004;13:390–407. doi: 10.1002/pon.761. [DOI] [PubMed] [Google Scholar]
  40. Pekarik E G, Prinz R J, Liebert D E, Weintraub S, Neale J M. The Pupil Evaluation Inventory: A sociometric technique for assessing children's social behavior. Journal of Abnormal Child Psychology. 1976;4:83–97. doi: 10.1007/BF00917607. [DOI] [PubMed] [Google Scholar]
  41. Prchal A, Landolt M A. How siblings of pediatric cancer patients experience the first time after diagnosis: A qualitative study. Cancer Nursing. 2012;35:133–140. doi: 10.1097/NCC.0b013e31821e0c59. [DOI] [PubMed] [Google Scholar]
  42. Rubin K H, Bukowski W M, Parker J G. 2006. Peer interactions, relationships and groups. In W. Damon (Series Ed.) and N. Eisenberg (Volume Ed.), The handbook of child psychology (6th edn. pp. 571–645). New York: Wiley. [Google Scholar]
  43. Sharpe D, Rossiter L. Siblings of children with a chronic illness: A meta-analysis. Journal of Pediatric Psychology. 2002;27:699–710. doi: 10.1093/jpepsy/27.8.699. [DOI] [PubMed] [Google Scholar]
  44. Sidhu R, Passmore A, Baker D. An investigation into parent perceptions of needs of siblings of children with cancer. Journal of Pediatric Oncology Nursing. 2005;22:276–287. doi: 10.1177/1043454205278480. [DOI] [PubMed] [Google Scholar]
  45. Sloper P. Experiences and support needs of siblings of children with cancer. Health and Social Care in the Community. 2000;8:298–306. doi: 10.1046/j.1365-2524.2000.00254.x. [DOI] [PubMed] [Google Scholar]
  46. Stocker C. Children's perceptions of relationships with siblings, friends, and mothers: Compensatory processes and links with adjustment. Journal of Child Psychology, Psychiatry and Allied Disciplines. 1994;35:1447–1459. doi: 10.1111/j.1469-7610.1994.tb01286.x. [DOI] [PubMed] [Google Scholar]
  47. Tabachnick B, Fidell L. Using multivariate statistics. New York: Harper Collins; 1989. [Google Scholar]
  48. Teja S, Stolberg A. Peer support, divorce, and child adjustment. Journal of Divorce and Remarriage. 1993;20:45–64. [Google Scholar]
  49. Vermaes I P R, van Susante A M J, van Bakel J A. Psychological functioning of siblings in families of children with chronic health conditions: A meta-analysis. Journal of Pediatric Psychology. 2012;37:166–184. doi: 10.1093/jpepsy/jsr081. [DOI] [PubMed] [Google Scholar]
  50. Vrijmoet-Wiersma C, van Klink J, Kolk A, Koopman H M, Ball L M, Maarten Egeler R. Assessment of parental psychological stress in pediatric cancer: A review. Journal of Pediatric Psychology. 2008;33:694–706. doi: 10.1093/jpepsy/jsn007. [DOI] [PubMed] [Google Scholar]
  51. Zeller M, Vannatta K, Schafer J, Noll R B. Behavioral reputation: A cross-age perspective. Developmental Psychology. 2003;39:129–139. doi: 10.1037//0012-1649.39.1.129. [DOI] [PubMed] [Google Scholar]

Articles from Journal of Pediatric Psychology are provided here courtesy of Oxford University Press

RESOURCES